PLS-SEM en R. Guía
Felipe A. López
julio de 2023
NOTA: Se entiende que usted maneja los conceptos básicos de Ecuaciones estructurales y que realizó la limpieza y validación de sus datos.
A. Carga de libreria y directorio a trabajar
A.1 Carga de librerias
# install.packages(pkgs = 'seminr')
# install.packages(“xlsx”)
#install.packages("genpathmox")
#install.packages("cSEM")
#install.packages("psych")
library(seminr)
library(xlsx)
library(cSEM)
library(genpathmox)
library('psych')A.2 Carga de datos
A.2.1 Carga de directorio de trabajo y datos
Reemplace directorio
getwd()## [1] "P:/R_Proyect/PLS-SEM/Proyecto/Rmark"directorio <- "P:/R_Proyect/PLS-SEM/Proyecto" ## Reemplace por su directorio
#setwd(directorio) # si desea dejar fijo el directorio de trabajo
getwd()## [1] "P:/R_Proyect/PLS-SEM/Proyecto/Rmark"En file sustituya por archivo de datos
pls_data <- read.csv(file = "P:/R_Proyect/PLS-SEM/Proyecto/2023.TRI_MGA.csv", header = TRUE, sep = ';')
dim(pls_data) ## Ver cantidad de filas y columnas## [1] 383 54Ver cabecera de los datos y tipos
head(pls_data) ### Primeros datosstr(pls_data) ### Tipo de datos## 'data.frame': 383 obs. of 54 variables:
## $ ï..ID : int 1 2 3 4 5 6 7 8 9 10 ...
## $ PE1 : int 5 5 5 2 3 5 5 4 4 3 ...
## $ PE2 : int 4 5 5 2 4 5 5 5 4 4 ...
## $ PE3 : int 5 5 5 4 3 5 5 5 5 5 ...
## $ PE4 : int 4 5 4 3 3 4 1 1 5 5 ...
## $ EE1 : int 3 3 4 2 2 5 3 4 4 5 ...
## $ EE2 : int 2 3 4 1 1 5 5 4 5 5 ...
## $ EE3 : int 3 3 2 1 1 5 5 5 5 5 ...
## $ SI1 : int 5 4 5 3 4 4 5 5 5 5 ...
## $ SI2 : int 4 4 5 3 4 3 4 4 4 5 ...
## $ SI3 : int 5 4 5 3 4 5 4 4 4 5 ...
## $ SI4 : int 4 4 5 3 1 4 5 5 4 5 ...
## $ FC1 : int 3 3 4 2 2 4 5 5 5 5 ...
## $ FC2 : int 1 3 2 1 1 5 5 5 3 5 ...
## $ FC3 : int 5 4 5 2 3 5 5 5 5 5 ...
## $ HM1 : int 4 5 5 3 3 5 5 5 5 5 ...
## $ HM2 : int 4 5 5 4 4 5 5 5 5 5 ...
## $ HM3 : int 4 5 5 3 4 5 5 5 5 5 ...
## $ HA1 : int 3 4 5 2 2 5 5 5 5 5 ...
## $ HA2 : int 3 5 5 1 2 5 4 5 5 5 ...
## $ HA3 : int 2 5 5 1 2 5 5 5 5 5 ...
## $ HA4 : int 2 4 5 1 2 4 5 5 5 5 ...
## $ HA5 : int 2 4 3 1 2 5 5 5 4 4 ...
## $ IU1 : int 4 5 5 3 3 5 5 5 5 5 ...
## $ IU2 : int 4 5 5 2 3 3 3 5 5 5 ...
## $ U1 : int 4 5 5 4 3 5 5 5 5 5 ...
## $ U2 : int 3 5 5 3 2 5 4 3 5 5 ...
## $ U3 : int 1 4 4 1 1 5 5 5 5 5 ...
## $ U4 : int 1 4 4 1 1 2 1 1 1 1 ...
## $ TRI1 : int 4 4 4 4 3 5 5 5 5 5 ...
## $ TRI2 : int 4 3 3 3 3 4 5 5 5 5 ...
## $ TRI3 : int 4 4 2 3 3 4 3 4 5 5 ...
## $ TRI4 : int 4 4 2 3 4 4 3 3 4 3 ...
## $ TRI5 : int 2 4 4 2 2 4 1 1 3 4 ...
## $ TRI6 : int 4 2 1 2 1 2 1 1 2 1 ...
## $ TRI7 : int 2 2 1 1 1 2 4 4 3 3 ...
## $ TRI8 : int 1 2 2 3 1 3 3 3 2 3 ...
## $ TRI9 : int 4 4 4 4 5 2 3 3 2 2 ...
## $ TRI10 : int 5 3 4 4 4 2 3 3 3 3 ...
## $ TRI11 : int 5 4 4 4 5 2 5 5 3 2 ...
## $ TRI12 : int 5 4 4 4 4 2 5 5 2 3 ...
## $ TRI13 : int 4 2 2 4 4 2 5 5 5 5 ...
## $ TRI14 : int 3 4 2 4 4 3 5 5 5 5 ...
## $ TRI15 : int 4 5 4 4 4 4 5 5 5 5 ...
## $ TRI16 : int 5 3 2 5 5 2 5 5 5 5 ...
## $ EXP : int 4 10 10 5 6 17 7 5 2 12 ...
## $ EDU : int 3 3 3 3 3 4 2 3 3 4 ...
## $ SOC : int 3 2 2 3 2 3 3 3 2 3 ...
## $ WSTATUS : chr "N" "N" "Y" "N" ...
## $ RETIRED : chr "Y" "Y" "N" "Y" ...
## $ GENDER : chr "Female" "Male" "Female" "Female" ...
## $ BORN : int 1943 1952 1954 1935 1935 1960 1949 1948 1957 1954 ...
## $ GENERATION: chr "Silent generation " "Early Baby boomer " "Early Baby boomer " "Silent generation " ...
## $ REGION : chr "BiobÃo" "BiobÃo" "BiobÃo" "BiobÃo" ...nrow(pls_data) ## numero filas## [1] 383ncol(pls_data) ## numero Columnas## [1] 54Crearemos una copia de la tabla en la que haremos los cambios
pls_data2 <-pls_dataA.2.2 Corrección de datos
Cambiar nombre a una variable
names(pls_data2)## [1] "ï..ID" "PE1" "PE2" "PE3" "PE4"
## [6] "EE1" "EE2" "EE3" "SI1" "SI2"
## [11] "SI3" "SI4" "FC1" "FC2" "FC3"
## [16] "HM1" "HM2" "HM3" "HA1" "HA2"
## [21] "HA3" "HA4" "HA5" "IU1" "IU2"
## [26] "U1" "U2" "U3" "U4" "TRI1"
## [31] "TRI2" "TRI3" "TRI4" "TRI5" "TRI6"
## [36] "TRI7" "TRI8" "TRI9" "TRI10" "TRI11"
## [41] "TRI12" "TRI13" "TRI14" "TRI15" "TRI16"
## [46] "EXP" "EDU" "SOC" "WSTATUS" "RETIRED"
## [51] "GENDER" "BORN" "GENERATION" "REGION"names(pls_data2)[1] = 'indice'Corregir nombre de la Región Bío-Bío
table(pls_data2[,54])##
## BiobÃo Coquimbo
## 259 124table(pls_data2$REGION)##
## BiobÃo Coquimbo
## 259 124pls_data2$REGION =ifelse(pls_data2$REGION=='BiobÃo', 'Bio-Bio', pls_data2$REGION)B. Estadistica descriptiva
B.1 Gráficos tablas
Crear tabla de frecuencia con variable categóricas
table(pls_data2$EDU)##
## 1 2 3 4
## 3 26 129 225tab1 <- table(pls_data2$EDU)
head(tab1)##
## 1 2 3 4
## 3 26 129 225barplot(tab1,
main = "Cantidad de datos por niveles de enseñanza",
xlab = "Nivel de enseñanza",
ylab = "Cantidad",
col = c("red", "green", "blue", 'yellow'),
)
table(pls_data2$SOC)##
## 1 2 3 4 5
## 7 55 253 67 1tab2 <- table(pls_data2$SOC)
head(tab2)##
## 1 2 3 4 5
## 7 55 253 67 1barplot(tab2,
main = "Cantidad de datos por Estado Civil",
xlab = "Estado civil",
ylab = "Cantidad",
col = c("red", "green", "blue", 'yellow', 'brown', 'orange'),
)
table(pls_data2$GENDER)##
## Female Male
## 213 170tab3 <- table(pls_data2$GENDER)
head(tab3)##
## Female Male
## 213 170barplot(tab3,
main = "Cantidad de datos por Género",
xlab = "Género",
ylab = "Cantidad",
col = c("red", "green", "blue", 'yellow', 'brown', 'orange'),
)
table(pls_data2$GENERATION)##
## Early Baby boomer Late Baby boomer Silent generation
## 128 160 95tab4 <- table(pls_data2$GENERATION)
head(tab4)##
## Early Baby boomer Late Baby boomer Silent generation
## 128 160 95barplot(tab4,
main = "Cantidad de datos por Generación",
xlab = "Generación",
ylab = "Cantidad",
col = c("red", "green", "blue", 'yellow', 'brown', 'orange'),
)
tab4c <- table(pls_data2$REGION)
barplot(tab4c, main = "Cantidad de datos por Región",
xlab = "Región", ylab = "Frecuencia", col = rainbow(2))
porcentaje <- round(tab4 / sum(tab4) * 100, 2)
colores <- rainbow(length(tab4))
pie(porcentaje, labels = paste0(porcentaje, "%"), main = "Porcentaje de Generación", col = colores)
legend("right", legend = names(tab4), cex = 0.8, fill = colores)
B.1.2 Pruebas normalidad
boxplot(pls_data2$BORN, main = "Gráfico de cajas Año nacimiento",
outline = TRUE)
hist(pls_data2$BORN, main = "Histograma Año nacimiento",
xlab = "Año de nacimiento",
ylab = "Frecuencia",
col = "red",
border = "black")
densidad_BORN <- density(pls_data$BORN)
plot(densidad_BORN,
main = "Densidad Experiencia Internet",
xlab = "Años Exp",
ylab = "Densidad")
skew(pls_data2$BORN,) # Simetría## [1] -0.7942247kurtosi(pls_data2$BORN,)## [1] -0.1702803multi.hist(pls_data2$BORN,dcol= c("blue","red"),dlty=c("dotted", "solid")) 
# Test de Shapiro
shapiro.test(pls_data2$BORN)##
## Shapiro-Wilk normality test
##
## data: pls_data2$BORN
## W = 0.90399, p-value = 7.724e-15# Test de kolmogorov-smirnov
ks.test(pls_data2$BORN, "pnorm", mean(pls_data2$BORN), sd(pls_data2$BORN))## Warning in ks.test(pls_data2$BORN, "pnorm", mean(pls_data2$BORN),
## sd(pls_data2$BORN)): ties should not be present for the Kolmogorov-Smirnov test##
## One-sample Kolmogorov-Smirnov test
##
## data: pls_data2$BORN
## D = 0.16525, p-value = 1.645e-09
## alternative hypothesis: two-sided#Con los siguientes comandos se pueden realizar pruebas adicionales de normalidad
#requiere paquete nortest
# require(nortest)
# ad.test(pls_data2$BORN) #test de Anderson-Darling
# cvm.test(pls_data2$BORN) #test de Cramer von mises
# pearson.test(pls_data2$BORN) #Chi cuadrado de pearsonboxplot(pls_data2$EXP, main = "Gráfico de cajas Años de Experiencia en Internet",
outline = TRUE)
densidad_EXP <- density(pls_data$EXP)
plot(densidad_EXP,
main = "Densidad Experiencia Internet",
xlab = "Años Exp",
ylab = "Densidad")
skew(pls_data$EXP) # Simetría## [1] -0.09126799kurtosi(pls_data$EXP)## [1] 0.04841059B.1.3 Otros gráficos
tabla1 <- table(pls_data2$REGION, pls_data2$GENDER)
barplot(tabla1,
main = "Gráfico por Género y Región",
xlab = "Género", ylab = "Frecuencia",
legend.text = rownames(tabla1),
beside = TRUE,
col = rainbow(2), label = TRUE)## Warning in plot.window(xlim, ylim, log = log, ...): "label" is not a graphical
## parameter## Warning in axis(if (horiz) 2 else 1, at = at.l, labels = names.arg, lty =
## axis.lty, : "label" is not a graphical parameter## Warning in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...): "label"
## is not a graphical parameter
tabla2 <- table(pls_data2$GENERATION, pls_data2$GENDER)
mosaicplot(tabla2, main = "Mosaico de Género y edad",
color = TRUE)
tabla3 <- table(pls_data2$SOC, pls_data2$GENERATION)
barplot(tabla3,
main = "Gráfico por Generación y Nivel socioeconomico",
xlab = "Generación", ylab = "Frecuencia",
legend.text = rownames(tabla3),
beside = FALSE,
col = rainbow(5), label = TRUE)## Warning in plot.window(xlim, ylim, log = log, ...): "label" is not a graphical
## parameter## Warning in axis(if (horiz) 2 else 1, at = at.l, labels = names.arg, lty =
## axis.lty, : "label" is not a graphical parameter## Warning in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...): "label"
## is not a graphical parameter
B.2 Separar variables categóricas
names(pls_data) ### Ver los nombres de las columnas ## [1] "ï..ID" "PE1" "PE2" "PE3" "PE4"
## [6] "EE1" "EE2" "EE3" "SI1" "SI2"
## [11] "SI3" "SI4" "FC1" "FC2" "FC3"
## [16] "HM1" "HM2" "HM3" "HA1" "HA2"
## [21] "HA3" "HA4" "HA5" "IU1" "IU2"
## [26] "U1" "U2" "U3" "U4" "TRI1"
## [31] "TRI2" "TRI3" "TRI4" "TRI5" "TRI6"
## [36] "TRI7" "TRI8" "TRI9" "TRI10" "TRI11"
## [41] "TRI12" "TRI13" "TRI14" "TRI15" "TRI16"
## [46] "EXP" "EDU" "SOC" "WSTATUS" "RETIRED"
## [51] "GENDER" "BORN" "GENERATION" "REGION"str(pls_data) ### Tipo de datos## 'data.frame': 383 obs. of 54 variables:
## $ ï..ID : int 1 2 3 4 5 6 7 8 9 10 ...
## $ PE1 : int 5 5 5 2 3 5 5 4 4 3 ...
## $ PE2 : int 4 5 5 2 4 5 5 5 4 4 ...
## $ PE3 : int 5 5 5 4 3 5 5 5 5 5 ...
## $ PE4 : int 4 5 4 3 3 4 1 1 5 5 ...
## $ EE1 : int 3 3 4 2 2 5 3 4 4 5 ...
## $ EE2 : int 2 3 4 1 1 5 5 4 5 5 ...
## $ EE3 : int 3 3 2 1 1 5 5 5 5 5 ...
## $ SI1 : int 5 4 5 3 4 4 5 5 5 5 ...
## $ SI2 : int 4 4 5 3 4 3 4 4 4 5 ...
## $ SI3 : int 5 4 5 3 4 5 4 4 4 5 ...
## $ SI4 : int 4 4 5 3 1 4 5 5 4 5 ...
## $ FC1 : int 3 3 4 2 2 4 5 5 5 5 ...
## $ FC2 : int 1 3 2 1 1 5 5 5 3 5 ...
## $ FC3 : int 5 4 5 2 3 5 5 5 5 5 ...
## $ HM1 : int 4 5 5 3 3 5 5 5 5 5 ...
## $ HM2 : int 4 5 5 4 4 5 5 5 5 5 ...
## $ HM3 : int 4 5 5 3 4 5 5 5 5 5 ...
## $ HA1 : int 3 4 5 2 2 5 5 5 5 5 ...
## $ HA2 : int 3 5 5 1 2 5 4 5 5 5 ...
## $ HA3 : int 2 5 5 1 2 5 5 5 5 5 ...
## $ HA4 : int 2 4 5 1 2 4 5 5 5 5 ...
## $ HA5 : int 2 4 3 1 2 5 5 5 4 4 ...
## $ IU1 : int 4 5 5 3 3 5 5 5 5 5 ...
## $ IU2 : int 4 5 5 2 3 3 3 5 5 5 ...
## $ U1 : int 4 5 5 4 3 5 5 5 5 5 ...
## $ U2 : int 3 5 5 3 2 5 4 3 5 5 ...
## $ U3 : int 1 4 4 1 1 5 5 5 5 5 ...
## $ U4 : int 1 4 4 1 1 2 1 1 1 1 ...
## $ TRI1 : int 4 4 4 4 3 5 5 5 5 5 ...
## $ TRI2 : int 4 3 3 3 3 4 5 5 5 5 ...
## $ TRI3 : int 4 4 2 3 3 4 3 4 5 5 ...
## $ TRI4 : int 4 4 2 3 4 4 3 3 4 3 ...
## $ TRI5 : int 2 4 4 2 2 4 1 1 3 4 ...
## $ TRI6 : int 4 2 1 2 1 2 1 1 2 1 ...
## $ TRI7 : int 2 2 1 1 1 2 4 4 3 3 ...
## $ TRI8 : int 1 2 2 3 1 3 3 3 2 3 ...
## $ TRI9 : int 4 4 4 4 5 2 3 3 2 2 ...
## $ TRI10 : int 5 3 4 4 4 2 3 3 3 3 ...
## $ TRI11 : int 5 4 4 4 5 2 5 5 3 2 ...
## $ TRI12 : int 5 4 4 4 4 2 5 5 2 3 ...
## $ TRI13 : int 4 2 2 4 4 2 5 5 5 5 ...
## $ TRI14 : int 3 4 2 4 4 3 5 5 5 5 ...
## $ TRI15 : int 4 5 4 4 4 4 5 5 5 5 ...
## $ TRI16 : int 5 3 2 5 5 2 5 5 5 5 ...
## $ EXP : int 4 10 10 5 6 17 7 5 2 12 ...
## $ EDU : int 3 3 3 3 3 4 2 3 3 4 ...
## $ SOC : int 3 2 2 3 2 3 3 3 2 3 ...
## $ WSTATUS : chr "N" "N" "Y" "N" ...
## $ RETIRED : chr "Y" "Y" "N" "Y" ...
## $ GENDER : chr "Female" "Male" "Female" "Female" ...
## $ BORN : int 1943 1952 1954 1935 1935 1960 1949 1948 1957 1954 ...
## $ GENERATION: chr "Silent generation " "Early Baby boomer " "Early Baby boomer " "Silent generation " ...
## $ REGION : chr "BiobÃo" "BiobÃo" "BiobÃo" "BiobÃo" ...categoricas <- c( 'EDU', 'SOC', 'WSTATUS', 'RETIRED', 'GENDER', 'GENERATION', 'REGION' )
otras <- c("PE1" , "PE2" , "PE3" , "PE4" , "EE1" , "EE2" , "EE3" , "SI1" , "SI2" , "SI3",
"SI4" , "FC1" , "FC2" , "FC3" , "HM1", "HM2" , "HM3" , "HA1" , "HA2" , "HA3" , "HA4",
"HA5" , "IU1" , "IU2" , "U1" , "U2" , "U3" , "U4" , "TRI1" , "TRI2" , "TRI3", "TRI4" ,
"TRI5" , "TRI6" , "TRI7" , "TRI8" , "TRI9" , "TRI10" ,"TRI11" ,"TRI12" , "TRI13" , "TRI14", "TRI15",
"TRI16")B.3 Generar resumen de campos no categóricos
resumen <- summary(pls_data2[,otras])
print(resumen)## PE1 PE2 PE3 PE4 EE1
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.00
## 1st Qu.:3.000 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:3.00
## Median :4.000 Median :4.000 Median :4.000 Median :4.000 Median :4.00
## Mean :3.943 Mean :4.125 Mean :4.407 Mean :4.266 Mean :3.41
## 3rd Qu.:5.000 3rd Qu.:5.000 3rd Qu.:5.000 3rd Qu.:5.000 3rd Qu.:4.00
## Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.00
## EE2 EE3 SI1 SI2
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:3.000 1st Qu.:3.000 1st Qu.:4.000 1st Qu.:4.000
## Median :4.000 Median :4.000 Median :4.000 Median :4.000
## Mean :3.366 Mean :3.376 Mean :4.225 Mean :4.172
## 3rd Qu.:4.000 3rd Qu.:4.000 3rd Qu.:5.000 3rd Qu.:5.000
## Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.000
## SI3 SI4 FC1 FC2 FC3
## Min. :1.000 Min. :1.000 Min. :1.00 Min. :1.000 Min. :1.000
## 1st Qu.:4.000 1st Qu.:3.000 1st Qu.:4.00 1st Qu.:3.000 1st Qu.:4.000
## Median :4.000 Median :4.000 Median :4.00 Median :4.000 Median :4.000
## Mean :4.141 Mean :3.809 Mean :3.99 Mean :3.577 Mean :4.269
## 3rd Qu.:5.000 3rd Qu.:5.000 3rd Qu.:5.00 3rd Qu.:4.000 3rd Qu.:5.000
## Max. :5.000 Max. :5.000 Max. :5.00 Max. :5.000 Max. :5.000
## HM1 HM2 HM3 HA1 HA2
## Min. :2.000 Min. :2.000 Min. :2.000 Min. :1.000 Min. :1.00
## 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:3.000 1st Qu.:2.00
## Median :4.000 Median :4.000 Median :4.000 Median :4.000 Median :4.00
## Mean :4.021 Mean :4.052 Mean :3.984 Mean :3.543 Mean :3.36
## 3rd Qu.:5.000 3rd Qu.:4.000 3rd Qu.:5.000 3rd Qu.:4.000 3rd Qu.:4.00
## Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.00
## HA3 HA4 HA5 IU1
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:3.000 1st Qu.:2.000 1st Qu.:2.000 1st Qu.:4.000
## Median :4.000 Median :3.000 Median :4.000 Median :4.000
## Mean :3.593 Mean :3.112 Mean :3.355 Mean :4.358
## 3rd Qu.:4.000 3rd Qu.:4.000 3rd Qu.:4.000 3rd Qu.:5.000
## Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.000
## IU2 U1 U2 U3 U4
## Min. :1.000 Min. :1.000 Min. :1.00 Min. :1.000 Min. :1.000
## 1st Qu.:3.000 1st Qu.:3.000 1st Qu.:3.00 1st Qu.:3.000 1st Qu.:1.000
## Median :4.000 Median :4.000 Median :4.00 Median :3.000 Median :2.000
## Mean :3.969 Mean :3.961 Mean :3.94 Mean :3.366 Mean :2.352
## 3rd Qu.:5.000 3rd Qu.:5.000 3rd Qu.:5.00 3rd Qu.:4.000 3rd Qu.:3.000
## Max. :5.000 Max. :5.000 Max. :5.00 Max. :5.000 Max. :5.000
## TRI1 TRI2 TRI3 TRI4 TRI5
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.00
## 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:3.000 1st Qu.:3.000 1st Qu.:2.00
## Median :4.000 Median :4.000 Median :4.000 Median :4.000 Median :3.00
## Mean :4.084 Mean :4.029 Mean :3.869 Mean :3.791 Mean :2.71
## 3rd Qu.:5.000 3rd Qu.:5.000 3rd Qu.:5.000 3rd Qu.:5.000 3rd Qu.:4.00
## Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.00
## TRI6 TRI7 TRI8 TRI9 TRI10
## Min. :1.000 Min. :1.000 Min. :1.00 Min. :1.000 Min. :1.000
## 1st Qu.:2.000 1st Qu.:2.000 1st Qu.:2.00 1st Qu.:2.000 1st Qu.:3.000
## Median :2.000 Median :3.000 Median :3.00 Median :4.000 Median :4.000
## Mean :2.326 Mean :2.875 Mean :3.18 Mean :3.261 Mean :3.452
## 3rd Qu.:3.000 3rd Qu.:4.000 3rd Qu.:4.00 3rd Qu.:4.000 3rd Qu.:4.000
## Max. :5.000 Max. :5.000 Max. :5.00 Max. :5.000 Max. :5.000
## TRI11 TRI12 TRI13 TRI14
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:3.000 1st Qu.:3.000 1st Qu.:3.000 1st Qu.:3.500
## Median :4.000 Median :4.000 Median :4.000 Median :4.000
## Mean :3.376 Mean :3.554 Mean :3.841 Mean :3.901
## 3rd Qu.:4.000 3rd Qu.:4.000 3rd Qu.:5.000 3rd Qu.:4.000
## Max. :5.000 Max. :5.000 Max. :5.000 Max. :5.000
## TRI15 TRI16
## Min. :1.000 Min. :1.000
## 1st Qu.:3.000 1st Qu.:2.000
## Median :4.000 Median :4.000
## Mean :3.883 Mean :3.308
## 3rd Qu.:5.000 3rd Qu.:4.000
## Max. :5.000 Max. :5.000Exportar a Excel con datos resumen
write.xlsx2(x=resumen,
'resumen.xlsx',
sheetName = "resumen",
col.names = TRUE,
row.names = TRUE,
append = FALSE,
showNA = TRUE,
password = NULL)B.4 Tablas de frecuencias
xtabs(~EDU + GENDER, data =pls_data2) ## Educacion y Género## GENDER
## EDU Female Male
## 1 2 1
## 2 17 9
## 3 79 50
## 4 115 110xtabs(~GENDER + WSTATUS, data =pls_data2) ##Género y Estatus Laboral## WSTATUS
## GENDER N Y
## Female 128 85
## Male 63 107xtabs(~GENDER + RETIRED, data =pls_data2) ##Género y Retirado## RETIRED
## GENDER N Y
## Female 70 143
## Male 80 90xtabs(~GENDER + GENERATION, data =pls_data2) ## Género y Generación## GENERATION
## GENDER Early Baby boomer Late Baby boomer Silent generation
## Female 72 83 58
## Male 56 77 37xtabs(~GENDER + REGION, data =pls_data2) ##Género y Región## REGION
## GENDER Bio-Bio Coquimbo
## Female 145 68
## Male 114 56xtabs(~REGION + GENERATION, data =pls_data2) ##Género y Región## GENERATION
## REGION Early Baby boomer Late Baby boomer Silent generation
## Bio-Bio 87 109 63
## Coquimbo 41 51 32C. Datos Faltantes
C.1 Obtener columnas con datos faltantes
Si aparece list() no hay datos faltantes
nan <- function(df) {
nulos <- list()
for (i in 1:length(df)) {
if (sum(is.na(df[[i]])) != 0) {
nulos[[length(nulos) + 1]] <- c(names(df)[i], sum(is.na(df[[i]])))
}
}
print(nulos)
}
nan(pls_data)## list()C.2 Reemplazar datos faltantes por -99
Si cambia a un valor distinto, luego al estimar modelo cambiar.
reemp_falt <- function(df) {
for (i in 1:length(df)) {
if (sum(is.na(df[[i]])) != 0) {
df[[i]] <- replace(df[[i]], is.na(df[[i]]), -99)
}
}
return(df)
}
pls_data2 <-reemp_falt(pls_data2)C.3 Eliminar datos faltantes o con una condicion
Eliminar los que se desea
pls_data2$PE1 == "-99"## [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [13] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [25] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [37] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [49] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [61] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [73] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [85] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [97] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [109] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [121] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [133] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [145] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [157] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [169] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [181] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [193] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [205] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [217] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [229] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [241] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [253] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [265] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [277] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [289] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [301] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [313] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [325] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [337] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [349] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [361] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
## [373] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSEpls_data2 <- pls_data2[(pls_data2$PE1 != "-99"),]D. Crear una variable categórica desde otra variable
D.1 Convertir en categoricas
str(pls_data2)## 'data.frame': 383 obs. of 54 variables:
## $ indice : int 1 2 3 4 5 6 7 8 9 10 ...
## $ PE1 : int 5 5 5 2 3 5 5 4 4 3 ...
## $ PE2 : int 4 5 5 2 4 5 5 5 4 4 ...
## $ PE3 : int 5 5 5 4 3 5 5 5 5 5 ...
## $ PE4 : int 4 5 4 3 3 4 1 1 5 5 ...
## $ EE1 : int 3 3 4 2 2 5 3 4 4 5 ...
## $ EE2 : int 2 3 4 1 1 5 5 4 5 5 ...
## $ EE3 : int 3 3 2 1 1 5 5 5 5 5 ...
## $ SI1 : int 5 4 5 3 4 4 5 5 5 5 ...
## $ SI2 : int 4 4 5 3 4 3 4 4 4 5 ...
## $ SI3 : int 5 4 5 3 4 5 4 4 4 5 ...
## $ SI4 : int 4 4 5 3 1 4 5 5 4 5 ...
## $ FC1 : int 3 3 4 2 2 4 5 5 5 5 ...
## $ FC2 : int 1 3 2 1 1 5 5 5 3 5 ...
## $ FC3 : int 5 4 5 2 3 5 5 5 5 5 ...
## $ HM1 : int 4 5 5 3 3 5 5 5 5 5 ...
## $ HM2 : int 4 5 5 4 4 5 5 5 5 5 ...
## $ HM3 : int 4 5 5 3 4 5 5 5 5 5 ...
## $ HA1 : int 3 4 5 2 2 5 5 5 5 5 ...
## $ HA2 : int 3 5 5 1 2 5 4 5 5 5 ...
## $ HA3 : int 2 5 5 1 2 5 5 5 5 5 ...
## $ HA4 : int 2 4 5 1 2 4 5 5 5 5 ...
## $ HA5 : int 2 4 3 1 2 5 5 5 4 4 ...
## $ IU1 : int 4 5 5 3 3 5 5 5 5 5 ...
## $ IU2 : int 4 5 5 2 3 3 3 5 5 5 ...
## $ U1 : int 4 5 5 4 3 5 5 5 5 5 ...
## $ U2 : int 3 5 5 3 2 5 4 3 5 5 ...
## $ U3 : int 1 4 4 1 1 5 5 5 5 5 ...
## $ U4 : int 1 4 4 1 1 2 1 1 1 1 ...
## $ TRI1 : int 4 4 4 4 3 5 5 5 5 5 ...
## $ TRI2 : int 4 3 3 3 3 4 5 5 5 5 ...
## $ TRI3 : int 4 4 2 3 3 4 3 4 5 5 ...
## $ TRI4 : int 4 4 2 3 4 4 3 3 4 3 ...
## $ TRI5 : int 2 4 4 2 2 4 1 1 3 4 ...
## $ TRI6 : int 4 2 1 2 1 2 1 1 2 1 ...
## $ TRI7 : int 2 2 1 1 1 2 4 4 3 3 ...
## $ TRI8 : int 1 2 2 3 1 3 3 3 2 3 ...
## $ TRI9 : int 4 4 4 4 5 2 3 3 2 2 ...
## $ TRI10 : int 5 3 4 4 4 2 3 3 3 3 ...
## $ TRI11 : int 5 4 4 4 5 2 5 5 3 2 ...
## $ TRI12 : int 5 4 4 4 4 2 5 5 2 3 ...
## $ TRI13 : int 4 2 2 4 4 2 5 5 5 5 ...
## $ TRI14 : int 3 4 2 4 4 3 5 5 5 5 ...
## $ TRI15 : int 4 5 4 4 4 4 5 5 5 5 ...
## $ TRI16 : int 5 3 2 5 5 2 5 5 5 5 ...
## $ EXP : int 4 10 10 5 6 17 7 5 2 12 ...
## $ EDU : int 3 3 3 3 3 4 2 3 3 4 ...
## $ SOC : int 3 2 2 3 2 3 3 3 2 3 ...
## $ WSTATUS : chr "N" "N" "Y" "N" ...
## $ RETIRED : chr "Y" "Y" "N" "Y" ...
## $ GENDER : chr "Female" "Male" "Female" "Female" ...
## $ BORN : int 1943 1952 1954 1935 1935 1960 1949 1948 1957 1954 ...
## $ GENERATION: chr "Silent generation " "Early Baby boomer " "Early Baby boomer " "Silent generation " ...
## $ REGION : chr "Bio-Bio" "Bio-Bio" "Bio-Bio" "Bio-Bio" ...pls_data2$EDU3=pls_data2$EDU
pls_data2$SOC3= pls_data2$SOC
pls_data2$EXP3= pls_data2$EXP
pls_data2$EDU2= as.factor(pls_data2$EDU)
pls_data2$SOC2= as.factor(pls_data2$SOC)
pls_data2$EXP2= as.factor(pls_data2$EXP)D.2 Cambiar variables categóricas
pls_data2$GENERO= ifelse(pls_data2$GENDER=='Male', 1, 2)
pls_data2$REGION3= ifelse(pls_data2$REGION=='Coquimbo', 1, 2)
pls_data2$WSTATUS3= ifelse(pls_data2$WSTATUS=='N', 1, 2)
pls_data2$RETIRED3= ifelse(pls_data2$RETIRED=='N', 1, 2)
pls_data2$GENERATION3= ifelse(pls_data2$GENERATION=="Silent generation ", 1,pls_data2$GENERATION)
pls_data2$GENERATION3= ifelse(pls_data2$GENERATION=="Late Baby boomer ", 3, pls_data2$GENERATION3)
pls_data2$GENERATION3= ifelse(pls_data2$GENERATION=="Early Baby boomer ", 2,pls_data2$GENERATION3)
pls_data2$GENERATION3 <- as.numeric(pls_data2$GENERATION3)
#pls_data2$WSTATUS2= as.factor(pls_data2$WSTATUS)
#pls_data2$REGION2= as.factor(pls_data2$REGION)
pls_data2$GENERO3= pls_data2$GENERO
#pls_data2$GENERATION2= as.factor(pls_data2$GENERATION)
#pls_data2$RETIRED2 = as.factor(pls_data2$RETIRED)E. Modelo de ecuaciones estructuales (semir)
E.1 Crear el modelo de medida
Por defecto se crean como reflectivo, para crear formativo agregar “weights = mode_B”
E.1.1 PLS Normal
## Reflectivo = mode_A (default)
## Formativo = mode_B (weights = mode_B)
modelo_medida <- constructs(
composite('PE', multi_items('PE', 1:4), weights = mode_A),
composite('EE', multi_items('EE', 1:3)),
composite('SI', multi_items('SI', 1:4)),
composite('FC', multi_items('FC', 1:3)),
composite('HM', multi_items('HM', 1:3)),
composite('HA', multi_items('HA', 1:5)),
# composite('CUSA', single_item('cusa')), # En el caso de ser un unico item dejar como single_item
composite('IU', multi_items('IU', 1:2)),
composite('SNS', multi_items('U', 1:4))
)
plot(modelo_medida)save_plot("modelo_medida.pdf")E.1.2 PLS Consistente
modelo_medida <- constructs(
reflective('PE', multi_items('PE', 1:4)),
reflective('EE', multi_items('EE', 1:3)),
reflective('SI', multi_items('SI', 1:4)),
reflective('FC', multi_items('FC', 1:3)),
reflective('HM', multi_items('HM', 1:3)),
reflective('HA', multi_items('HA', 1:5)),
# composite('CUSA', single_item('cusa')), # En el caso de ser un unico item dejar como single_item
reflective('IU', multi_items('IU', 1:2)),
reflective('SNS', multi_items('U', 1:4))
)
plot(modelo_medida)save_plot("modelo_medida.pdf")E.2 Crear Modelo estructural
modelo_estruc <- relationships(
paths(from = c('PE', 'EE', 'SI', 'FC', 'HM', "HA"), to = c('IU')),
paths(from = c('FC', 'HA', "IU"), to = c('SNS'))
)
## ----- Generamos el modelo con colores
thm <- seminr_theme_create(plot.rounding = 2, ## Decimales
plot.adj = FALSE,
sm.node.fill = "cadetblue1",
mm.node.fill = "lightgray",
)
seminr_theme_set(thm)
## ----
plot(modelo_estruc, title = "Fig. 1: Modelo Estructural")save_plot("fig1.Modelo_Estructural.pdf")E.3 Estimación del Modelo
estimacion_model <- estimate_pls(data = pls_data2,
measurement_model = modelo_medida, #Constructos
structural_model = modelo_estruc, # Caminos Path
inner_weights = path_weighting,
# path_weighting para path weighting (default) o path_factorial para factor weighting,
missing = mean_replacement, #Reemplazar los valores perdido mean es default
missing_value = '-99' ) # Valores perdidos
summary_estimacion_model <- summary(estimacion_model)
plot(estimacion_model, title = "Fig. 2: Modelo Estimado")save_plot("fig2.Modelo_Estimado.pdf")E.4 Reportes modelo
E.4.1. Valores perdidos y estadisticas de cada variable
summary_estimacion_model$descriptives$statistics ## Valores perdidos y representación ## $items
## No. Missing Mean Median Min Max Std.Dev. Kurtosis Skewness
## PE1 1.000 0.000 3.943 4.000 1.000 5.000 0.969 3.276 -0.853
## PE2 2.000 0.000 4.125 4.000 1.000 5.000 0.877 4.645 -1.178
## PE3 3.000 0.000 4.407 4.000 1.000 5.000 0.652 5.589 -1.157
## PE4 4.000 0.000 4.266 4.000 1.000 5.000 0.692 5.197 -0.976
## EE1 5.000 0.000 3.410 4.000 1.000 5.000 1.119 2.373 -0.508
## EE2 6.000 0.000 3.366 4.000 1.000 5.000 1.105 2.362 -0.455
## EE3 7.000 0.000 3.376 4.000 1.000 5.000 1.095 2.397 -0.473
## SI1 8.000 0.000 4.225 4.000 1.000 5.000 0.757 4.261 -0.978
## SI2 9.000 0.000 4.172 4.000 1.000 5.000 0.750 4.039 -0.852
## SI3 10.000 0.000 4.141 4.000 1.000 5.000 0.803 4.673 -1.079
## SI4 11.000 0.000 3.809 4.000 1.000 5.000 1.118 2.884 -0.780
## FC1 12.000 0.000 3.990 4.000 1.000 5.000 0.850 3.698 -0.877
## FC2 13.000 0.000 3.577 4.000 1.000 5.000 1.043 2.815 -0.711
## FC3 14.000 0.000 4.269 4.000 1.000 5.000 0.677 6.450 -1.196
## HM1 15.000 0.000 4.021 4.000 2.000 5.000 0.765 2.954 -0.491
## HM2 16.000 0.000 4.052 4.000 2.000 5.000 0.703 3.590 -0.570
## HM3 17.000 0.000 3.984 4.000 2.000 5.000 0.812 2.924 -0.559
## HA1 18.000 0.000 3.543 4.000 1.000 5.000 1.113 2.126 -0.422
## HA2 19.000 0.000 3.360 4.000 1.000 5.000 1.213 1.897 -0.214
## HA3 20.000 0.000 3.593 4.000 1.000 5.000 1.098 2.209 -0.462
## HA4 21.000 0.000 3.112 3.000 1.000 5.000 1.264 1.790 0.084
## HA5 22.000 0.000 3.355 4.000 1.000 5.000 1.230 1.931 -0.285
## IU1 23.000 0.000 4.358 4.000 1.000 5.000 0.663 6.294 -1.194
## IU2 24.000 0.000 3.969 4.000 1.000 5.000 0.968 2.978 -0.771
## U1 25.000 0.000 3.961 4.000 1.000 5.000 1.112 3.005 -0.895
## U2 26.000 0.000 3.940 4.000 1.000 5.000 1.002 2.945 -0.661
## U3 27.000 0.000 3.366 3.000 1.000 5.000 1.262 2.244 -0.344
## U4 28.000 0.000 2.352 2.000 1.000 5.000 1.267 2.086 0.468
##
## $constructs
## No. Missing Mean Median Min Max Std.Dev. Kurtosis Skewness
## PE 1.000 0.000 0.000 -0.159 -3.712 1.239 1.000 3.223 -0.596
## EE 2.000 0.000 -0.000 0.283 -2.302 1.558 1.000 2.491 -0.531
## SI 3.000 0.000 0.000 -0.141 -2.967 1.272 1.000 2.512 -0.309
## FC 4.000 0.000 0.000 -0.005 -4.152 1.514 1.000 4.501 -0.795
## HM 5.000 0.000 0.000 -0.029 -2.893 1.403 1.000 2.912 -0.346
## HA 6.000 0.000 -0.000 0.034 -2.255 1.504 1.000 2.047 -0.119
## IU 7.000 0.000 0.000 -0.286 -4.490 1.116 1.000 4.035 -0.789
## SNS 8.000 0.000 0.000 -0.081 -2.779 1.721 1.000 2.274 -0.105x <- summary_estimacion_model$descriptives$statistics
write.xlsx2(x=x["items"],
'resumen.xlsx',
sheetName = "resumen_hor",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)
write.xlsx2(x=x["constructs"],
'resumen.xlsx',
sheetName = "resumen_hor_const",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.2. Número de iteraciones
Nota: Si es mayor a 300 significa que no converge
summary_estimacion_model$iterations ## [1] 5E.4.3. R^2
Exogenos
summary_estimacion_model$paths ## IU SNS
## R^2 0.772 0.816
## AdjR^2 0.768 0.814
## PE 0.375 .
## EE -0.190 .
## SI 0.198 .
## FC 0.171 0.079
## HM 0.108 .
## HA 0.317 0.636
## IU . 0.257plot(summary_estimacion_model$paths[,1], pch = 2, col = "red", main="Betas y R^2 (Exogenos)",
xlab = "Variables", ylab = "Valores estimados", xlim = c(0,length(row.names(summary_estimacion_model$paths))+1)
)
text(summary_estimacion_model$paths[,1],labels = row.names(summary_estimacion_model$paths) , pos = 4)
Endogenos
plot(summary_estimacion_model$paths[,2], pch = 2, col = "red", main="Betas y R^2 (Endogenos)",
xlab = "Variables", ylab = "Valores estimados" , xlim = c(0,length(row.names(summary_estimacion_model$paths))+1) )
text(summary_estimacion_model$paths[,2],labels = row.names(summary_estimacion_model$paths) , pos = 4)
Exportar Excel
write.xlsx2(x=summary_estimacion_model$paths,
'resumen.xlsx',
sheetName = "BetasyR",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.4. Fiabilidad
Cronbach’s alpha (alpha), composite reliability (rhoC), average variance extracted (AVE),
summary_estimacion_model$reliability ## alpha rhoC AVE rhoA
## PE 0.831 0.832 0.556 0.840
## EE 0.930 0.930 0.818 0.947
## SI 0.864 0.857 0.602 0.864
## FC 0.716 0.712 0.455 0.724
## HM 0.910 0.910 0.771 0.911
## HA 0.942 0.942 0.765 0.943
## IU 0.794 0.794 0.659 0.795
## SNS 0.769 0.771 0.459 0.776
##
## Alpha, rhoC, and rhoA should exceed 0.7 while AVE should exceed 0.5plot(summary_estimacion_model$reliability, title = "Fig. 3: Fiabilidad")
Alpha
plot(summary_estimacion_model$reliability[,1], pch = 1, col = "red", main="Alpha ",
xlab = "Variables", ylab = "Valor estimados", ylim = c(0, 1))
text(summary_estimacion_model$reliability[,1],labels = row.names(summary_estimacion_model$reliability) , pos = 3)
abline(h=0.7,col="red",lty=2,lwd=2) 
AVE
plot(summary_estimacion_model$reliability[,3], pch = 2, col = "red", main="AVE ",
xlab = "Variables", ylab = "Valor estimados", ylim = c(0, 1))
text(summary_estimacion_model$reliability[,1],labels = row.names(summary_estimacion_model$reliability) , pos = 3)
abline(h=0.5,col="red",lty=2,lwd=2) 
Exportar a Excel
write.xlsx2(x=summary_estimacion_model$reliability,
'resumen.xlsx',
sheetName = "reliability",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.5. Cargas
summary_estimacion_model$loadings # Cargas -> reflectivas mayor a 0.70## PE EE SI FC HM HA IU SNS
## PE1 0.643 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE2 0.840 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE3 0.720 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE4 0.764 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## EE1 0.000 1.040 0.000 0.000 0.000 0.000 0.000 0.000
## EE2 0.000 0.863 0.000 0.000 0.000 0.000 0.000 0.000
## EE3 0.000 0.793 0.000 0.000 0.000 0.000 0.000 0.000
## SI1 0.000 0.000 0.734 0.000 0.000 0.000 0.000 0.000
## SI2 0.000 0.000 0.702 0.000 0.000 0.000 0.000 0.000
## SI3 0.000 0.000 0.779 0.000 0.000 0.000 0.000 0.000
## SI4 0.000 0.000 0.879 0.000 0.000 0.000 0.000 0.000
## FC1 0.000 0.000 0.000 0.558 0.000 0.000 0.000 0.000
## FC2 0.000 0.000 0.000 0.699 0.000 0.000 0.000 0.000
## FC3 0.000 0.000 0.000 0.752 0.000 0.000 0.000 0.000
## HM1 0.000 0.000 0.000 0.000 0.898 0.000 0.000 0.000
## HM2 0.000 0.000 0.000 0.000 0.846 0.000 0.000 0.000
## HM3 0.000 0.000 0.000 0.000 0.890 0.000 0.000 0.000
## HA1 0.000 0.000 0.000 0.000 0.000 0.883 0.000 0.000
## HA2 0.000 0.000 0.000 0.000 0.000 0.884 0.000 0.000
## HA3 0.000 0.000 0.000 0.000 0.000 0.834 0.000 0.000
## HA4 0.000 0.000 0.000 0.000 0.000 0.880 0.000 0.000
## HA5 0.000 0.000 0.000 0.000 0.000 0.892 0.000 0.000
## IU1 0.000 0.000 0.000 0.000 0.000 0.000 0.830 0.000
## IU2 0.000 0.000 0.000 0.000 0.000 0.000 0.793 0.000
## U1 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.675
## U2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.651
## U3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.762
## U4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.613summary_estimacion_model$loadings^2## PE EE SI FC HM HA IU SNS
## PE1 0.413 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE2 0.706 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE3 0.518 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE4 0.584 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## EE1 0.000 1.081 0.000 0.000 0.000 0.000 0.000 0.000
## EE2 0.000 0.745 0.000 0.000 0.000 0.000 0.000 0.000
## EE3 0.000 0.628 0.000 0.000 0.000 0.000 0.000 0.000
## SI1 0.000 0.000 0.538 0.000 0.000 0.000 0.000 0.000
## SI2 0.000 0.000 0.492 0.000 0.000 0.000 0.000 0.000
## SI3 0.000 0.000 0.606 0.000 0.000 0.000 0.000 0.000
## SI4 0.000 0.000 0.772 0.000 0.000 0.000 0.000 0.000
## FC1 0.000 0.000 0.000 0.311 0.000 0.000 0.000 0.000
## FC2 0.000 0.000 0.000 0.489 0.000 0.000 0.000 0.000
## FC3 0.000 0.000 0.000 0.566 0.000 0.000 0.000 0.000
## HM1 0.000 0.000 0.000 0.000 0.806 0.000 0.000 0.000
## HM2 0.000 0.000 0.000 0.000 0.715 0.000 0.000 0.000
## HM3 0.000 0.000 0.000 0.000 0.791 0.000 0.000 0.000
## HA1 0.000 0.000 0.000 0.000 0.000 0.779 0.000 0.000
## HA2 0.000 0.000 0.000 0.000 0.000 0.781 0.000 0.000
## HA3 0.000 0.000 0.000 0.000 0.000 0.696 0.000 0.000
## HA4 0.000 0.000 0.000 0.000 0.000 0.774 0.000 0.000
## HA5 0.000 0.000 0.000 0.000 0.000 0.795 0.000 0.000
## IU1 0.000 0.000 0.000 0.000 0.000 0.000 0.689 0.000
## IU2 0.000 0.000 0.000 0.000 0.000 0.000 0.629 0.000
## U1 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.456
## U2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.424
## U3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.580
## U4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.376summary_estimacion_model$weights # Pesos -> Formativos## PE EE SI FC HM HA IU SNS
## PE1 0.265 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE2 0.347 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE3 0.297 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE4 0.315 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## EE1 0.000 0.412 0.000 0.000 0.000 0.000 0.000 0.000
## EE2 0.000 0.342 0.000 0.000 0.000 0.000 0.000 0.000
## EE3 0.000 0.314 0.000 0.000 0.000 0.000 0.000 0.000
## SI1 0.000 0.000 0.283 0.000 0.000 0.000 0.000 0.000
## SI2 0.000 0.000 0.271 0.000 0.000 0.000 0.000 0.000
## SI3 0.000 0.000 0.300 0.000 0.000 0.000 0.000 0.000
## SI4 0.000 0.000 0.339 0.000 0.000 0.000 0.000 0.000
## FC1 0.000 0.000 0.000 0.348 0.000 0.000 0.000 0.000
## FC2 0.000 0.000 0.000 0.436 0.000 0.000 0.000 0.000
## FC3 0.000 0.000 0.000 0.469 0.000 0.000 0.000 0.000
## HM1 0.000 0.000 0.000 0.000 0.370 0.000 0.000 0.000
## HM2 0.000 0.000 0.000 0.000 0.349 0.000 0.000 0.000
## HM3 0.000 0.000 0.000 0.000 0.367 0.000 0.000 0.000
## HA1 0.000 0.000 0.000 0.000 0.000 0.224 0.000 0.000
## HA2 0.000 0.000 0.000 0.000 0.000 0.224 0.000 0.000
## HA3 0.000 0.000 0.000 0.000 0.000 0.212 0.000 0.000
## HA4 0.000 0.000 0.000 0.000 0.000 0.223 0.000 0.000
## HA5 0.000 0.000 0.000 0.000 0.000 0.226 0.000 0.000
## IU1 0.000 0.000 0.000 0.000 0.000 0.000 0.562 0.000
## IU2 0.000 0.000 0.000 0.000 0.000 0.000 0.537 0.000
## U1 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.324
## U2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.313
## U3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.366
## U4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.294Exportar a Excel
write.xlsx2(x=summary_estimacion_model$loadings,
'resumen.xlsx',
sheetName = "loadings",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)
write.xlsx2(x=summary_estimacion_model$weights,
'resumen.xlsx',
sheetName = "weights",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.6. Cargas Cruzadas
summary_estimacion_model$validity$cross_loadings## PE EE SI FC HM HA IU SNS
## PE1 0.763 0.436 0.411 0.412 0.504 0.488 0.456 0.528
## PE2 0.870 0.450 0.457 0.471 0.556 0.499 0.596 0.559
## PE3 0.832 0.333 0.492 0.416 0.536 0.382 0.510 0.396
## PE4 0.789 0.388 0.468 0.465 0.493 0.429 0.542 0.427
## EE1 0.495 0.931 0.314 0.650 0.478 0.598 0.441 0.545
## EE2 0.470 0.953 0.247 0.622 0.441 0.560 0.366 0.475
## EE3 0.406 0.923 0.238 0.595 0.433 0.511 0.336 0.446
## SI1 0.442 0.235 0.860 0.343 0.360 0.328 0.450 0.371
## SI2 0.459 0.201 0.884 0.331 0.357 0.329 0.430 0.375
## SI3 0.480 0.180 0.857 0.360 0.401 0.400 0.477 0.381
## SI4 0.493 0.335 0.767 0.382 0.385 0.422 0.539 0.422
## FC1 0.398 0.478 0.272 0.778 0.467 0.481 0.361 0.341
## FC2 0.388 0.741 0.278 0.806 0.435 0.547 0.393 0.485
## FC3 0.501 0.381 0.445 0.807 0.465 0.431 0.491 0.455
## HM1 0.624 0.455 0.404 0.543 0.919 0.576 0.579 0.549
## HM2 0.565 0.468 0.435 0.521 0.912 0.595 0.545 0.504
## HM3 0.579 0.415 0.404 0.508 0.930 0.595 0.574 0.545
## HA1 0.543 0.573 0.353 0.552 0.628 0.896 0.573 0.684
## HA2 0.528 0.546 0.402 0.524 0.570 0.935 0.557 0.699
## HA3 0.441 0.600 0.376 0.645 0.599 0.855 0.567 0.626
## HA4 0.485 0.492 0.450 0.505 0.560 0.904 0.571 0.682
## HA5 0.483 0.491 0.423 0.519 0.526 0.913 0.568 0.700
## IU1 0.603 0.377 0.511 0.524 0.566 0.580 0.915 0.573
## IU2 0.580 0.374 0.528 0.433 0.555 0.566 0.906 0.534
## U1 0.422 0.411 0.342 0.333 0.431 0.584 0.483 0.755
## U2 0.447 0.373 0.346 0.423 0.487 0.528 0.522 0.766
## U3 0.529 0.422 0.398 0.475 0.499 0.666 0.482 0.831
## U4 0.393 0.418 0.339 0.441 0.357 0.529 0.381 0.722write.xlsx2(x=summary_estimacion_model$validity$cross_loadings,
'resumen.xlsx',
sheetName = "cross_loadings",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.7. VIF
summary_estimacion_model$vif_antecedents## IU :
## PE EE SI FC HM HA
## 2.184 2.087 1.566 2.283 2.200 2.237
##
## SNS :
## FC HA IU
## 1.678 2.008 1.752summary_estimacion_model$validity$vif_items ## PE :
## PE1 PE2 PE3 PE4
## 1.736 2.260 1.995 1.675
##
## EE :
## EE1 EE2 EE3
## 3.128 5.439 4.161
##
## SI :
## SI1 SI2 SI3 SI4
## 2.858 3.256 2.271 1.455
##
## FC :
## FC1 FC2 FC3
## 1.477 1.426 1.336
##
## HM :
## HM1 HM2 HM3
## 2.950 2.912 3.340
##
## HA :
## HA1 HA2 HA3 HA4 HA5
## 3.490 5.116 2.560 3.779 3.967
##
## IU :
## IU1 IU2
## 1.764 1.764
##
## SNS :
## U1 U2 U3 U4
## 1.459 1.501 1.708 1.422E.4.8. Fornell-Larcker
summary_estimacion_model$validity$fl_criteria## PE EE SI FC HM HA IU SNS
## PE 0.745 . . . . . . .
## EE 0.493 0.905 . . . . . .
## SI 0.561 0.289 0.776 . . . . .
## FC 0.542 0.668 0.424 0.675 . . . .
## HM 0.641 0.484 0.450 0.570 0.878 . . .
## HA 0.551 0.599 0.445 0.608 0.639 0.875 . .
## IU 0.650 0.413 0.570 0.527 0.616 0.630 0.812 .
## SNS 0.586 0.527 0.464 0.544 0.579 0.753 0.608 0.678
##
## FL Criteria table reports square root of AVE on the diagonal and construct correlations on the lower triangle.write.xlsx2(x=summary_estimacion_model$validity$fl_criteria,
'resumen.xlsx',
sheetName = "Fornell-Larcker",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.9. fSquare
summary_estimacion_model$fSquare ## PE EE SI FC HM HA IU SNS
## PE 0.000 0.000 0.000 0.000 0.000 0.000 0.196 0.000
## EE 0.000 0.000 0.000 0.000 0.000 0.000 0.049 0.000
## SI 0.000 0.000 0.000 0.000 0.000 0.000 0.090 0.000
## FC 0.000 0.000 0.000 0.000 0.000 0.000 0.024 0.013
## HM 0.000 0.000 0.000 0.000 0.000 0.000 0.018 0.000
## HA 0.000 0.000 0.000 0.000 0.000 0.000 0.169 0.782
## IU 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.149
## SNS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000write.xlsx2(x=summary_estimacion_model$fSquare,
'resumen.xlsx',
sheetName = "fSquare",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.10. HTMT
summary_estimacion_model$validity$htmt ## PE EE SI FC HM HA IU SNS
## PE . . . . . . . .
## EE 0.556 . . . . . . .
## SI 0.657 0.310 . . . . . .
## FC 0.695 0.815 0.523 . . . . .
## HM 0.737 0.524 0.504 0.707 . . . .
## HA 0.624 0.636 0.486 0.744 0.692 . . .
## IU 0.795 0.474 0.680 0.688 0.724 0.728 . .
## SNS 0.730 0.618 0.563 0.721 0.689 0.881 0.777 .write.xlsx2(x=summary_estimacion_model$validity$htmt ,
'resumen.xlsx',
sheetName = "htmt",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.11. Tabla de correlaciones
summary_estimacion_model$descriptives$correlations$constructs ## PE EE SI FC HM HA IU SNS
## PE 1.000 0.493 0.561 0.542 0.641 0.551 0.650 0.586
## EE 0.493 1.000 0.289 0.668 0.484 0.599 0.413 0.527
## SI 0.561 0.289 1.000 0.424 0.450 0.445 0.570 0.464
## FC 0.542 0.668 0.424 1.000 0.570 0.608 0.527 0.544
## HM 0.641 0.484 0.450 0.570 1.000 0.639 0.616 0.579
## HA 0.551 0.599 0.445 0.608 0.639 1.000 0.630 0.753
## IU 0.650 0.413 0.570 0.527 0.616 0.630 1.000 0.608
## SNS 0.586 0.527 0.464 0.544 0.579 0.753 0.608 1.000write.xlsx2(x=summary_estimacion_model$descriptives$correlations$constructs ,
'resumen.xlsx',
sheetName = "Correl_constructos",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.4.12. Otros
b) Efectos totales
c) Efectos indirectos
d) Puntuaciones estimadas para los constructos
e) seleccion de modelo BIC, AIC
summary_estimacion_model$total_effects ## b)## PE EE SI FC HM HA IU SNS
## PE 0.000 0.000 0.000 0.000 0.000 0.000 0.375 0.096
## EE 0.000 0.000 0.000 0.000 0.000 0.000 -0.190 -0.049
## SI 0.000 0.000 0.000 0.000 0.000 0.000 0.198 0.051
## FC 0.000 0.000 0.000 0.000 0.000 0.000 0.171 0.123
## HM 0.000 0.000 0.000 0.000 0.000 0.000 0.108 0.028
## HA 0.000 0.000 0.000 0.000 0.000 0.000 0.317 0.717
## IU 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.257
## SNS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000summary_estimacion_model$total_indirect_effects ## c)## PE EE SI FC HM HA IU SNS
## PE 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.096
## EE 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.049
## SI 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.051
## FC 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.044
## HM 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.028
## HA 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.082
## IU 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SNS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000# summary_estimacion_model$composite_scores ## d)
summary_estimacion_model$it_criteria ## e)## IU SNS
## AIC -320.517 -345.750
## BIC -292.881 -329.958E.5. Estimación Bootstrap
boot_estimacion <- bootstrap_model(seminr_model = estimacion_model , #modelo estimado E.3 estimate_pls()
nboot = 500, ### N° Subsamples >5000
cores = parallel::detectCores(), #CPU cores -parallel processing
seed = 123) #Semilla inicial
sum_boot <- summary(boot_estimacion,
alpha=0.05 ### Intervalo de confianza, en este caso es dos colas 90%
)
plot(boot_estimacion, title = "Fig. 4 Bootstrapped Model")save_plot("fig4.Bootstrapped_Modelo.pdf")E.6. Reportes Bootstrapped
E.6.1. Paths
sum_boot$bootstrapped_paths ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## PE -> IU 0.375 0.383 0.108 3.479 0.172 0.598
## EE -> IU -0.190 -0.198 0.150 -1.269 -0.493 0.030
## SI -> IU 0.198 0.192 0.095 2.079 0.026 0.369
## FC -> IU 0.171 0.182 0.250 0.684 -0.156 0.659
## FC -> SNS 0.079 0.078 0.093 0.850 -0.115 0.258
## HM -> IU 0.108 0.099 0.095 1.144 -0.070 0.265
## HA -> IU 0.317 0.319 0.083 3.843 0.161 0.469
## HA -> SNS 0.636 0.634 0.085 7.469 0.481 0.803
## IU -> SNS 0.257 0.259 0.083 3.109 0.100 0.427write.xlsx2(x=sum_boot$bootstrapped_paths ,
'resumen.xlsx',
sheetName = "bootstrapped_Coef_Path",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.6.2. Cargas, pesos y efectos totales del modelo
sum_boot$bootstrapped_loadings ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## PE1 -> PE 0.643 0.642 0.055 11.588 0.526 0.739
## PE2 -> PE 0.840 0.837 0.039 21.287 0.755 0.908
## PE3 -> PE 0.720 0.716 0.055 13.153 0.609 0.808
## PE4 -> PE 0.764 0.770 0.060 12.806 0.657 0.878
## EE1 -> EE 1.040 1.043 0.051 20.565 0.944 1.147
## EE2 -> EE 0.863 0.859 0.040 21.536 0.772 0.935
## EE3 -> EE 0.793 0.791 0.048 16.443 0.692 0.876
## SI1 -> SI 0.734 0.726 0.063 11.639 0.594 0.833
## SI2 -> SI 0.702 0.700 0.049 14.383 0.599 0.788
## SI3 -> SI 0.779 0.778 0.045 17.429 0.692 0.857
## SI4 -> SI 0.879 0.880 0.061 14.480 0.757 0.984
## FC1 -> FC 0.558 0.554 0.062 8.964 0.419 0.664
## FC2 -> FC 0.699 0.695 0.056 12.496 0.578 0.808
## FC3 -> FC 0.752 0.749 0.042 17.700 0.668 0.827
## HM1 -> HM 0.898 0.899 0.035 25.629 0.829 0.960
## HM2 -> HM 0.846 0.842 0.037 22.981 0.766 0.904
## HM3 -> HM 0.890 0.889 0.028 31.964 0.835 0.949
## HA1 -> HA 0.883 0.882 0.020 45.192 0.841 0.917
## HA2 -> HA 0.884 0.884 0.020 45.154 0.843 0.920
## HA3 -> HA 0.834 0.833 0.026 31.971 0.778 0.883
## HA4 -> HA 0.880 0.882 0.022 40.056 0.841 0.923
## HA5 -> HA 0.892 0.893 0.022 39.640 0.847 0.933
## IU1 -> IU 0.830 0.829 0.023 35.386 0.781 0.871
## IU2 -> IU 0.793 0.792 0.026 30.782 0.742 0.841
## U1 -> SNS 0.675 0.676 0.034 19.893 0.603 0.742
## U2 -> SNS 0.651 0.651 0.037 17.641 0.580 0.719
## U3 -> SNS 0.762 0.762 0.030 25.317 0.704 0.821
## U4 -> SNS 0.613 0.614 0.035 17.398 0.541 0.683sum_boot$bootstrapped_weights #bootstrap standard error, t-statistic, and confidence intervals for the indicator weights## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## PE1 -> PE 0.265 0.264 0.020 13.363 0.226 0.301
## PE2 -> PE 0.347 0.345 0.018 18.745 0.310 0.379
## PE3 -> PE 0.297 0.295 0.022 13.789 0.251 0.336
## PE4 -> PE 0.315 0.317 0.024 13.178 0.272 0.362
## EE1 -> EE 0.412 0.414 0.023 17.642 0.374 0.467
## EE2 -> EE 0.342 0.341 0.014 24.769 0.313 0.367
## EE3 -> EE 0.314 0.314 0.017 18.921 0.281 0.344
## SI1 -> SI 0.283 0.281 0.021 13.229 0.238 0.321
## SI2 -> SI 0.271 0.270 0.014 19.790 0.242 0.295
## SI3 -> SI 0.300 0.301 0.015 19.496 0.270 0.333
## SI4 -> SI 0.339 0.341 0.029 11.742 0.289 0.398
## FC1 -> FC 0.348 0.347 0.026 13.476 0.293 0.393
## FC2 -> FC 0.436 0.436 0.030 14.407 0.381 0.502
## FC3 -> FC 0.469 0.471 0.032 14.453 0.406 0.538
## HM1 -> HM 0.370 0.372 0.015 24.502 0.343 0.402
## HM2 -> HM 0.349 0.348 0.012 29.951 0.322 0.369
## HM3 -> HM 0.367 0.367 0.012 30.609 0.346 0.391
## HA1 -> HA 0.224 0.224 0.005 42.536 0.214 0.235
## HA2 -> HA 0.224 0.224 0.005 49.239 0.215 0.233
## HA3 -> HA 0.212 0.211 0.006 34.659 0.200 0.223
## HA4 -> HA 0.223 0.224 0.005 42.076 0.213 0.234
## HA5 -> HA 0.226 0.226 0.005 42.781 0.216 0.237
## IU1 -> IU 0.562 0.562 0.013 44.223 0.537 0.588
## IU2 -> IU 0.537 0.537 0.011 46.956 0.516 0.561
## U1 -> SNS 0.324 0.324 0.014 23.155 0.297 0.351
## U2 -> SNS 0.313 0.312 0.013 23.632 0.289 0.340
## U3 -> SNS 0.366 0.365 0.016 23.054 0.337 0.395
## U4 -> SNS 0.294 0.294 0.015 19.186 0.264 0.325sum_boot$bootstrapped_total_paths #bootstrap standard error, t-statistic, and confidence intervals total effects## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## PE -> IU 0.375 0.383 0.108 3.479 0.172 0.598
## PE -> SNS 0.096 0.099 0.042 2.296 0.027 0.188
## EE -> IU -0.190 -0.198 0.150 -1.269 -0.493 0.030
## EE -> SNS -0.049 -0.050 0.043 -1.127 -0.122 0.007
## SI -> IU 0.198 0.192 0.095 2.079 0.026 0.369
## SI -> SNS 0.051 0.050 0.030 1.684 0.004 0.110
## FC -> IU 0.171 0.182 0.250 0.684 -0.156 0.659
## FC -> SNS 0.123 0.124 0.115 1.072 -0.077 0.337
## HM -> IU 0.108 0.099 0.095 1.144 -0.070 0.265
## HM -> SNS 0.028 0.026 0.027 1.014 -0.018 0.080
## HA -> IU 0.317 0.319 0.083 3.843 0.161 0.469
## HA -> SNS 0.717 0.717 0.079 9.129 0.576 0.880
## IU -> SNS 0.257 0.259 0.083 3.109 0.100 0.427write.xlsx2(x=sum_boot$bootstrapped_loadings ,
'resumen.xlsx',
sheetName = "bootstrapped_loadings",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)
write.xlsx2(x=sum_boot$bootstrapped_weights ,
'resumen.xlsx',
sheetName = "bootstrapped_weights",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)
write.xlsx2(x=sum_boot$bootstrapped_total_paths ,
'resumen.xlsx',
sheetName = "bootstrapped_total_paths",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)E.6.3. HTMT CI
sum_boot$bootstrapped_HTMT ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## PE -> EE 0.556 0.556 0.045 12.385 0.464 0.641
## PE -> SI 0.657 0.660 0.048 13.782 0.568 0.753
## PE -> FC 0.695 0.698 0.052 13.367 0.594 0.790
## PE -> HM 0.737 0.737 0.035 21.122 0.670 0.800
## PE -> HA 0.624 0.625 0.040 15.430 0.540 0.698
## PE -> IU 0.795 0.796 0.038 20.780 0.723 0.864
## PE -> SNS 0.730 0.731 0.048 15.137 0.636 0.817
## EE -> SI 0.310 0.311 0.055 5.681 0.198 0.414
## EE -> FC 0.815 0.815 0.035 23.482 0.740 0.878
## EE -> HM 0.524 0.522 0.045 11.628 0.437 0.600
## EE -> HA 0.636 0.637 0.034 18.793 0.575 0.702
## EE -> IU 0.474 0.473 0.050 9.402 0.367 0.569
## EE -> SNS 0.618 0.617 0.040 15.340 0.534 0.687
## SI -> FC 0.523 0.531 0.064 8.182 0.401 0.645
## SI -> HM 0.504 0.506 0.049 10.219 0.412 0.601
## SI -> HA 0.486 0.484 0.047 10.344 0.385 0.576
## SI -> IU 0.680 0.680 0.048 14.067 0.581 0.774
## SI -> SNS 0.563 0.563 0.050 11.173 0.461 0.661
## FC -> HM 0.707 0.708 0.041 17.348 0.618 0.786
## FC -> HA 0.744 0.746 0.043 17.333 0.659 0.825
## FC -> IU 0.688 0.689 0.043 16.039 0.609 0.784
## FC -> SNS 0.721 0.722 0.041 17.679 0.645 0.794
## HM -> HA 0.692 0.691 0.034 20.184 0.620 0.753
## HM -> IU 0.724 0.724 0.036 20.304 0.653 0.791
## HM -> SNS 0.689 0.689 0.042 16.256 0.607 0.775
## HA -> IU 0.728 0.729 0.041 17.779 0.643 0.805
## HA -> SNS 0.881 0.880 0.023 37.923 0.833 0.921
## IU -> SNS 0.777 0.777 0.042 18.695 0.696 0.858summary_estimacion_model$validity$htmt ### HTMT modelo estructural ( <0.85 )## PE EE SI FC HM HA IU SNS
## PE . . . . . . . .
## EE 0.556 . . . . . . .
## SI 0.657 0.310 . . . . . .
## FC 0.695 0.815 0.523 . . . . .
## HM 0.737 0.524 0.504 0.707 . . . .
## HA 0.624 0.636 0.486 0.744 0.692 . . .
## IU 0.795 0.474 0.680 0.688 0.724 0.728 . .
## SNS 0.730 0.618 0.563 0.721 0.689 0.881 0.777 .write.xlsx2(x=sum_boot$bootstrapped_HTMT ,
'resumen.xlsx',
sheetName = "bootstrapped_HTMT",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)F. Análisis de Mediación
F.1 Efectos Indirectos
Efectos totales indirectos
summary_estimacion_model$total_indirect_effects## PE EE SI FC HM HA IU SNS
## PE 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.096
## EE 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.049
## SI 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.051
## FC 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.044
## HM 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.028
## HA 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.082
## IU 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SNS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000#Evaluación de la significancia de los efectos indirectos. p1 * p2 es significante
specific_effect_significance(boot_estimacion, ###Boot
from ='FC',
through = 'IU', ### podría ser un vertor del tipo c('construct1', 'construct2')).
to = 'SNS',
alpha = 0.05)## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## 0.04399887 0.04551061 0.07082786 0.62120854 -0.04255763
## 97.5% CI
## 0.17648413specific_effect_significance(boot_estimacion, ###Boot
from ='HA',
through = 'IU',
to = 'SNS',
alpha = 0.05)## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## 0.08155251 0.08312684 0.03662836 2.22648555 0.02400088
## 97.5% CI
## 0.16578543#FC ==> SNS No significativo ==> Evaluar si p3 es Directo o no efecto
#HA ==> SNS Significativo ==> Efecto Complementario/ Competitivo o Indirecto solo
sum_boot$total_indirect_effects## NULLF.2 Efecto directo
F.2.1 Paso 1: Significancia
Evaluar la significancia y luego para ver si es mediación full o parcial se revisan los path directos.
summary_estimacion_model$paths## IU SNS
## R^2 0.772 0.816
## AdjR^2 0.768 0.814
## PE 0.375 .
## EE -0.190 .
## SI 0.198 .
## FC 0.171 0.079
## HM 0.108 .
## HA 0.317 0.636
## IU . 0.257sum_boot$bootstrapped_paths ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## PE -> IU 0.375 0.383 0.108 3.479 0.172 0.598
## EE -> IU -0.190 -0.198 0.150 -1.269 -0.493 0.030
## SI -> IU 0.198 0.192 0.095 2.079 0.026 0.369
## FC -> IU 0.171 0.182 0.250 0.684 -0.156 0.659
## FC -> SNS 0.079 0.078 0.093 0.850 -0.115 0.258
## HM -> IU 0.108 0.099 0.095 1.144 -0.070 0.265
## HA -> IU 0.317 0.319 0.083 3.843 0.161 0.469
## HA -> SNS 0.636 0.634 0.085 7.469 0.481 0.803
## IU -> SNS 0.257 0.259 0.083 3.109 0.100 0.427#FC ==> SNS No significativo ==> No effecto
#HA ==> SNS Significativo ==> Evaluar si es complementario (0<) o competitivo (0>)F.2.2 Paso 2: tipo de mediación
## Calcula el signo de ESE CAMINO p1*p2*p3 complementario (0<) o competitivo (0>)
summary_estimacion_model$paths['HA', 'SNS'] *
summary_estimacion_model$paths['HA', 'IU'] *
summary_estimacion_model$paths['IU', 'SNS'] ## [1] 0.05182717summary_estimacion_model$paths['FC', 'SNS'] *
summary_estimacion_model$paths['FC', 'IU'] *
summary_estimacion_model$paths['IU', 'SNS'] ## [1] 0.003467245G. Predict PLS
G.1. Generar la predicción del modelo
predict_modelo <- predict_pls(
model = estimacion_model, ### modelo de medida E.3
technique = predict_DA,
# direct antecedent (predict_DA) consideraría tanto el antecedente como el mediador pedictor del constructo
# earliest antecedent (predict_EA) mediador se excluiría del análisis
noFolds = 10, ### Folds a generar
reps = 10) ### Numero de repeticiones CV
sum_predict_modelo <- summary(predict_modelo)
#sum_predict_modeloComparamos los RMSE de PLS out-of-sample metrics v/s LM out-of-sample metrics. Si PLS<LM Ok
#sum_predict_modelo$PLS_out_of_sample
#sum_predict_modelo$LM_out_of_sample
predict_dif <- sum_predict_modelo$PLS_out_of_sample-sum_predict_modelo$LM_out_of_sample
predict_dif ## IU1 IU2 U1 U2 U3 U4
## RMSE 0.005 -0.021 0.022 0.028 0.011 -0.016
## MAE -0.008 -0.013 0.025 0.040 0.021 -0.007# Si todos los items son negativos ==> Alta predicción (PLS<LM)
# Si es la mayoría ==> Baja predicción
# Si ninguno ==> No poder de predicciónwrite.xlsx2(x=predict_dif,
'resumen.xlsx',
sheetName = "Predict_dif (PLS-LM)",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)sum_predict_modelo$prediction_errorwrite.xlsx2(x=sum_predict_modelo$prediction_error,
'resumen.xlsx',
sheetName = "Predict_erro",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)G.2. Analizar la distribución del error (indicador en específico)
par(mfrow=c(1,3))
plot(sum_predict_modelo,
indicator = 'IU1')
plot(sum_predict_modelo,
indicator = 'IU2')
par(mfrow=c(1,1))
par(mfrow=c(1,4))
plot(sum_predict_modelo,
indicator = 'U1')
plot(sum_predict_modelo,
indicator = 'U2')
plot(sum_predict_modelo,
indicator = 'U3')
plot(sum_predict_modelo,
indicator = 'U4')
par(mfrow=c(1,1))H. Análisis de Moderadores
H.1. Modelo de medida Moderadores
modelo_medida_mod <- constructs(
composite('PE', multi_items('PE', 1:4), weights = mode_A),
composite('EE', multi_items('EE', 1:3)),
composite('SI', multi_items('SI', 1:4)),
composite('FC', multi_items('FC', 1:3)),
composite('HM', multi_items('HM', 1:3)),
composite('HA', multi_items('HA', 1:5)),
# composite('CUSA', single_item('cusa')), # En el caso de ser un unico item dejar como single_item
composite('TRI_A', multi_items('TRI', 1:4)),
composite('TRI_B', multi_items('TRI', 5:8)),
composite('IU', multi_items('IU', 1:2)),
composite('SNS', multi_items('U', 1:4)),
interaction_term(iv = 'IU', moderator = 'TRI_A', method = two_stage), #Moderador method = orthogonal o method = two_stage
interaction_term(iv = 'FC', moderator = 'TRI_B', method = two_stage) #Moderador method = orthogonal o method = two_stage
)
plot(modelo_medida_mod)H.2. Modelo Estructural Moderadores
modelo_estruc_mod <- relationships(
paths(from = c('PE', 'EE', 'SI', 'HM','FC', "HA"), to = c('IU')),
paths(from = c('HA'), to = c('SNS')),
paths(from = c('IU', 'TRI_A', 'IU*TRI_A'), to = c('SNS')),
paths(from = c('FC', 'TRI_B', 'FC*TRI_B'), to = c('SNS'))
)
plot(modelo_estruc_mod)H.3. Ejecución modelo
pls_model_mod_med <- estimate_pls(data = pls_data2,
measurement_model = modelo_medida_mod,
structural_model = modelo_estruc_mod,
missing = mean_replacement,
missing_value = '-99'
)
boot_pls_model_mod_med <- bootstrap_model(seminr_model = pls_model_mod_med,
nboot = 500) #Cambiar al menos a 5000
sum_pls_model_mod_med <- summary(pls_model_mod_med)
sum_boot_pls_model_mod <- summary(boot_pls_model_mod_med, alpha = 0.05)
plot(pls_model_mod_med, title = "Fig. 5: Bootstrap Modelo Estimado Moderador")save_plot("fig 5.Bootstrap Modelo Estimado Moderador.pdf")H.4. Evaluar el modelo Moderador
H.4.1. R^2 Exogenos
sum_pls_model_mod_med$paths ## IU SNS
## R^2 0.581 0.613
## AdjR^2 0.575 0.606
## PE 0.269 .
## EE -0.085 .
## SI 0.209 .
## HM 0.159 .
## FC 0.084 0.066
## HA 0.286 0.531
## IU . 0.205
## TRI_A . -0.017
## IU*TRI_A . 0.021
## TRI_B . 0.119
## FC*TRI_B . 0.040plot(sum_pls_model_mod_med$paths[,1], pch = 2, col = "red", main="Betas y R^2 moderador (Exogenos)",
xlab = "Variables", ylab = "Valores estimados", xlim = c(0,length(row.names(sum_pls_model_mod_med$paths))+1)
)
text(sum_pls_model_mod_med$paths[,1],labels = row.names(sum_pls_model_mod_med$paths) , pos = 4)
Endogenos
plot(sum_pls_model_mod_med$paths[,2], pch = 2, col = "red", main="Betas y R^2 moderador (Endogenos)",
xlab = "Variables", ylab = "Valores estimados" , xlim = c(0,length(row.names(sum_pls_model_mod_med$paths))+1) )
text(sum_pls_model_mod_med$paths[,2],labels = row.names(sum_pls_model_mod_med$paths) , pos = 4)
Exportar Excel
write.xlsx2(x=sum_pls_model_mod_med$paths,
'resumen.xlsx',
sheetName = "BetasyR_Moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)H.4.2. Fiabilidad Cronbach’s alpha (alpha), composite reliability (rhoC), average variance extracted (AVE),
sum_pls_model_mod_med$reliability ## alpha rhoC AVE rhoA
## PE 0.831 0.887 0.664 0.840
## EE 0.930 0.955 0.876 0.947
## SI 0.864 0.907 0.711 0.864
## HM 0.910 0.943 0.847 0.911
## FC 0.716 0.839 0.635 0.724
## HA 0.942 0.956 0.812 0.943
## IU 0.794 0.906 0.829 0.795
## TRI_A 0.866 0.908 0.712 0.872
## IU*TRI_A 1.000 1.000 1.000 1.000
## TRI_B 0.816 0.877 0.642 0.832
## FC*TRI_B 1.000 1.000 1.000 1.000
## SNS 0.769 0.853 0.592 0.774
##
## Alpha, rhoC, and rhoA should exceed 0.7 while AVE should exceed 0.5plot(sum_pls_model_mod_med$reliability)
Exportar a Excel
write.xlsx2(x=sum_pls_model_mod_med$reliability,
'resumen.xlsx',
sheetName = "reliability_moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)H.4.3. Cargas
sum_pls_model_mod_med$loadings # Cargas -> reflectivas mayor a 0.70## PE EE SI HM FC HA IU TRI_A IU*TRI_A
## PE1 0.763 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.000
## PE2 0.870 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.000
## PE3 0.832 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE4 0.789 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## EE1 0.000 0.931 0.000 0.000 0.000 0.000 0.000 0.000 -0.000
## EE2 0.000 0.953 0.000 0.000 0.000 0.000 0.000 0.000 -0.000
## EE3 0.000 0.923 0.000 0.000 0.000 0.000 0.000 0.000 -0.000
## SI1 0.000 0.000 0.860 0.000 0.000 0.000 0.000 0.000 0.000
## SI2 0.000 0.000 0.884 0.000 0.000 0.000 0.000 0.000 0.000
## SI3 0.000 0.000 0.857 0.000 0.000 0.000 0.000 0.000 0.000
## SI4 0.000 0.000 0.767 0.000 0.000 0.000 0.000 0.000 0.000
## FC1 0.000 0.000 0.000 0.000 0.778 0.000 0.000 0.000 0.000
## FC2 0.000 0.000 0.000 0.000 0.806 0.000 0.000 0.000 -0.000
## FC3 0.000 0.000 0.000 0.000 0.807 0.000 0.000 0.000 0.000
## HM1 0.000 0.000 0.000 0.919 0.000 0.000 0.000 0.000 0.000
## HM2 0.000 0.000 0.000 0.912 0.000 0.000 0.000 0.000 0.000
## HM3 0.000 0.000 0.000 0.930 0.000 0.000 0.000 0.000 0.000
## HA1 0.000 0.000 0.000 0.000 0.000 0.896 0.000 0.000 0.000
## HA2 0.000 0.000 0.000 0.000 0.000 0.935 0.000 0.000 0.000
## HA3 0.000 0.000 0.000 0.000 0.000 0.855 0.000 0.000 0.000
## HA4 0.000 0.000 0.000 0.000 0.000 0.904 0.000 0.000 0.000
## HA5 0.000 0.000 0.000 0.000 0.000 0.913 0.000 0.000 0.000
## TRI1 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.810 0.000
## TRI2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.844 0.000
## TRI3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.884 0.000
## TRI4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.836 0.000
## TRI5 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.000
## TRI6 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.000
## TRI7 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## TRI8 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## IU1 0.000 0.000 0.000 0.000 0.000 0.000 0.915 0.000 -0.000
## IU2 0.000 0.000 0.000 0.000 0.000 0.000 0.906 0.000 -0.000
## U1 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## U2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.000
## U3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## U4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## IU*TRI_A_intxn -0.000 -0.000 0.000 0.000 0.000 0.000 -0.000 0.000 0.888
## FC*TRI_B_intxn -0.000 -0.000 0.000 -0.000 -0.000 -0.000 -0.000 -0.000 0.000
## TRI_B FC*TRI_B SNS
## PE1 0.000 -0.000 0.000
## PE2 0.000 -0.000 0.000
## PE3 0.000 -0.000 0.000
## PE4 0.000 -0.000 0.000
## EE1 0.000 -0.000 0.000
## EE2 0.000 -0.000 0.000
## EE3 0.000 -0.000 0.000
## SI1 0.000 0.000 0.000
## SI2 0.000 0.000 0.000
## SI3 0.000 0.000 0.000
## SI4 0.000 0.000 0.000
## FC1 0.000 -0.000 0.000
## FC2 0.000 -0.000 0.000
## FC3 0.000 -0.000 0.000
## HM1 0.000 -0.000 0.000
## HM2 0.000 -0.000 0.000
## HM3 0.000 -0.000 0.000
## HA1 0.000 -0.000 0.000
## HA2 0.000 -0.000 0.000
## HA3 0.000 -0.000 0.000
## HA4 0.000 -0.000 0.000
## HA5 0.000 -0.000 0.000
## TRI1 0.000 0.000 0.000
## TRI2 0.000 -0.000 0.000
## TRI3 0.000 -0.000 0.000
## TRI4 0.000 -0.000 0.000
## TRI5 0.800 -0.000 0.000
## TRI6 0.753 -0.000 0.000
## TRI7 0.856 -0.000 0.000
## TRI8 0.792 -0.000 0.000
## IU1 0.000 -0.000 0.000
## IU2 0.000 -0.000 0.000
## U1 0.000 -0.000 0.752
## U2 0.000 -0.000 0.765
## U3 0.000 -0.000 0.828
## U4 0.000 -0.000 0.729
## IU*TRI_A_intxn 0.000 0.000 0.000
## FC*TRI_B_intxn -0.000 1.132 -0.000sum_pls_model_mod_med$weights # Pesos -> Formativos## PE EE SI HM FC HA IU TRI_A IU*TRI_A TRI_B
## PE1 0.265 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE2 0.347 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE3 0.297 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## PE4 0.315 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## EE1 0.000 0.412 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## EE2 0.000 0.342 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## EE3 0.000 0.314 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SI1 0.000 0.000 0.283 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SI2 0.000 0.000 0.271 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SI3 0.000 0.000 0.300 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SI4 0.000 0.000 0.339 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC1 0.000 0.000 0.000 0.000 0.348 0.000 0.000 0.000 0.000 0.000
## FC2 0.000 0.000 0.000 0.000 0.436 0.000 0.000 0.000 0.000 0.000
## FC3 0.000 0.000 0.000 0.000 0.468 0.000 0.000 0.000 0.000 0.000
## HM1 0.000 0.000 0.000 0.370 0.000 0.000 0.000 0.000 0.000 0.000
## HM2 0.000 0.000 0.000 0.349 0.000 0.000 0.000 0.000 0.000 0.000
## HM3 0.000 0.000 0.000 0.367 0.000 0.000 0.000 0.000 0.000 0.000
## HA1 0.000 0.000 0.000 0.000 0.000 0.224 0.000 0.000 0.000 0.000
## HA2 0.000 0.000 0.000 0.000 0.000 0.224 0.000 0.000 0.000 0.000
## HA3 0.000 0.000 0.000 0.000 0.000 0.212 0.000 0.000 0.000 0.000
## HA4 0.000 0.000 0.000 0.000 0.000 0.223 0.000 0.000 0.000 0.000
## HA5 0.000 0.000 0.000 0.000 0.000 0.226 0.000 0.000 0.000 0.000
## TRI1 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.248 0.000 0.000
## TRI2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.312 0.000 0.000
## TRI3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.311 0.000 0.000
## TRI4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.311 0.000 0.000
## TRI5 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.328
## TRI6 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.213
## TRI7 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.362
## TRI8 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.338
## IU1 0.000 0.000 0.000 0.000 0.000 0.000 0.562 0.000 0.000 0.000
## IU2 0.000 0.000 0.000 0.000 0.000 0.000 0.537 0.000 0.000 0.000
## U1 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## U2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## U3 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## U4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## IU*TRI_A_intxn 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 0.000
## FC*TRI_B_intxn 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC*TRI_B SNS
## PE1 0.000 0.000
## PE2 0.000 0.000
## PE3 0.000 0.000
## PE4 0.000 0.000
## EE1 0.000 0.000
## EE2 0.000 0.000
## EE3 0.000 0.000
## SI1 0.000 0.000
## SI2 0.000 0.000
## SI3 0.000 0.000
## SI4 0.000 0.000
## FC1 0.000 0.000
## FC2 0.000 0.000
## FC3 0.000 0.000
## HM1 0.000 0.000
## HM2 0.000 0.000
## HM3 0.000 0.000
## HA1 0.000 0.000
## HA2 0.000 0.000
## HA3 0.000 0.000
## HA4 0.000 0.000
## HA5 0.000 0.000
## TRI1 0.000 0.000
## TRI2 0.000 0.000
## TRI3 0.000 0.000
## TRI4 0.000 0.000
## TRI5 0.000 0.000
## TRI6 0.000 0.000
## TRI7 0.000 0.000
## TRI8 0.000 0.000
## IU1 0.000 0.000
## IU2 0.000 0.000
## U1 0.000 0.322
## U2 0.000 0.313
## U3 0.000 0.358
## U4 0.000 0.305
## IU*TRI_A_intxn 0.000 0.000
## FC*TRI_B_intxn 1.000 0.000Exportar a Excel
write.xlsx2(x=sum_pls_model_mod_med$loadings,
'resumen.xlsx',
sheetName = "loadings_moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)
write.xlsx2(x=sum_pls_model_mod_med$weights,
'resumen.xlsx',
sheetName = "weights_moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)H.4.4. Cargas Cruzadas
sum_pls_model_mod_med$validity$cross_loadings## PE EE SI HM FC HA IU TRI_A IU*TRI_A
## PE1 0.763 0.436 0.411 0.504 0.412 0.488 0.456 0.509 -0.044
## PE2 0.870 0.450 0.457 0.556 0.471 0.499 0.596 0.532 -0.097
## PE3 0.832 0.333 0.492 0.536 0.416 0.382 0.510 0.440 0.001
## PE4 0.789 0.388 0.468 0.493 0.465 0.429 0.542 0.475 0.088
## EE1 0.495 0.931 0.314 0.478 0.650 0.598 0.441 0.421 -0.026
## EE2 0.470 0.953 0.247 0.441 0.622 0.560 0.366 0.402 -0.057
## EE3 0.406 0.923 0.238 0.433 0.595 0.511 0.336 0.399 -0.039
## SI1 0.442 0.235 0.860 0.360 0.343 0.328 0.450 0.346 0.139
## SI2 0.459 0.201 0.884 0.357 0.331 0.329 0.430 0.346 0.203
## SI3 0.480 0.180 0.857 0.401 0.359 0.400 0.477 0.381 0.166
## SI4 0.493 0.335 0.767 0.385 0.382 0.422 0.539 0.448 0.065
## FC1 0.398 0.478 0.272 0.467 0.778 0.481 0.361 0.367 0.105
## FC2 0.388 0.741 0.278 0.435 0.806 0.547 0.393 0.400 -0.021
## FC3 0.501 0.381 0.445 0.465 0.807 0.431 0.491 0.411 0.058
## HM1 0.624 0.455 0.404 0.919 0.543 0.576 0.579 0.556 0.027
## HM2 0.565 0.468 0.435 0.912 0.521 0.595 0.545 0.551 0.049
## HM3 0.579 0.415 0.404 0.930 0.508 0.595 0.574 0.539 0.028
## HA1 0.543 0.573 0.353 0.628 0.552 0.896 0.573 0.587 0.046
## HA2 0.528 0.546 0.402 0.570 0.524 0.935 0.557 0.589 0.101
## HA3 0.441 0.600 0.376 0.599 0.645 0.855 0.567 0.487 0.032
## HA4 0.485 0.492 0.450 0.560 0.505 0.904 0.571 0.611 0.135
## HA5 0.483 0.491 0.423 0.526 0.519 0.913 0.568 0.576 0.124
## TRI1 0.430 0.305 0.379 0.446 0.353 0.465 0.432 0.810 0.159
## TRI2 0.483 0.359 0.382 0.479 0.461 0.563 0.477 0.844 0.129
## TRI3 0.518 0.351 0.383 0.538 0.418 0.544 0.515 0.884 0.125
## TRI4 0.582 0.449 0.403 0.541 0.426 0.556 0.553 0.836 0.056
## TRI5 0.346 0.470 0.210 0.361 0.431 0.466 0.327 0.307 -0.009
## TRI6 0.287 0.409 0.159 0.241 0.334 0.288 0.233 0.229 -0.086
## TRI7 0.403 0.589 0.321 0.432 0.566 0.529 0.385 0.418 0.035
## TRI8 0.295 0.480 0.167 0.338 0.491 0.480 0.297 0.381 0.125
## IU1 0.603 0.377 0.511 0.566 0.524 0.580 0.915 0.506 -0.090
## IU2 0.580 0.374 0.528 0.555 0.433 0.566 0.906 0.568 -0.052
## U1 0.422 0.411 0.342 0.431 0.333 0.584 0.483 0.405 0.007
## U2 0.447 0.373 0.346 0.487 0.423 0.528 0.522 0.399 -0.035
## U3 0.529 0.422 0.398 0.499 0.475 0.666 0.482 0.496 0.120
## U4 0.393 0.418 0.339 0.357 0.441 0.529 0.381 0.307 0.123
## IU*TRI_A_intxn -0.017 -0.043 0.166 0.037 0.055 0.098 -0.078 0.136 1.000
## FC*TRI_B_intxn -0.084 -0.223 0.165 -0.130 -0.348 -0.158 -0.069 -0.021 0.259
## TRI_B FC*TRI_B SNS
## PE1 0.423 -0.093 0.527
## PE2 0.365 -0.109 0.558
## PE3 0.268 -0.062 0.395
## PE4 0.324 -0.009 0.428
## EE1 0.602 -0.209 0.545
## EE2 0.582 -0.221 0.475
## EE3 0.538 -0.195 0.446
## SI1 0.229 0.149 0.371
## SI2 0.225 0.189 0.375
## SI3 0.218 0.144 0.381
## SI4 0.247 0.083 0.422
## FC1 0.445 -0.306 0.342
## FC2 0.615 -0.292 0.486
## FC3 0.342 -0.243 0.455
## HM1 0.406 -0.123 0.548
## HM2 0.428 -0.146 0.503
## HM3 0.383 -0.091 0.545
## HA1 0.512 -0.169 0.683
## HA2 0.508 -0.128 0.699
## HA3 0.579 -0.244 0.626
## HA4 0.495 -0.045 0.682
## HA5 0.467 -0.130 0.700
## TRI1 0.281 0.052 0.370
## TRI2 0.339 -0.026 0.466
## TRI3 0.368 -0.041 0.465
## TRI4 0.448 -0.041 0.464
## TRI5 0.800 -0.019 0.437
## TRI6 0.753 -0.061 0.283
## TRI7 0.856 -0.099 0.482
## TRI8 0.792 -0.135 0.450
## IU1 0.354 -0.089 0.572
## IU2 0.369 -0.036 0.533
## U1 0.362 -0.038 0.752
## U2 0.418 -0.113 0.765
## U3 0.370 -0.067 0.828
## U4 0.493 -0.035 0.729
## IU*TRI_A_intxn 0.034 0.259 0.072
## FC*TRI_B_intxn -0.101 1.000 -0.082write.xlsx2(x=sum_pls_model_mod_med$validity$cross_loadings,
'resumen.xlsx',
sheetName = "cross_loadings_moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)H.4.5. VIF
sum_pls_model_mod_med$vif_antecedents## IU :
## PE EE SI HM FC HA
## 2.184 2.087 1.566 2.200 2.284 2.237
##
## SNS :
## HA IU TRI_A IU*TRI_A FC TRI_B FC*TRI_B
## 2.529 2.067 1.953 1.193 2.304 1.742 1.334sum_pls_model_mod_med$validity$vif_items ## PE :
## PE1 PE2 PE3 PE4
## 1.736 2.260 1.995 1.675
##
## EE :
## EE1 EE2 EE3
## 3.128 5.439 4.161
##
## SI :
## SI1 SI2 SI3 SI4
## 2.858 3.256 2.271 1.455
##
## HM :
## HM1 HM2 HM3
## 2.950 2.912 3.340
##
## FC :
## FC1 FC2 FC3
## 1.477 1.426 1.336
##
## HA :
## HA1 HA2 HA3 HA4 HA5
## 3.490 5.116 2.560 3.779 3.967
##
## IU :
## IU1 IU2
## 1.764 1.764
##
## TRI_A :
## TRI1 TRI2 TRI3 TRI4
## 1.967 2.076 2.563 2.065
##
## IU*TRI_A :
## IU*TRI_A_intxn
## 1
##
## TRI_B :
## TRI5 TRI6 TRI7 TRI8
## 1.785 1.746 1.930 1.636
##
## FC*TRI_B :
## FC*TRI_B_intxn
## 1
##
## SNS :
## U1 U2 U3 U4
## 1.459 1.501 1.708 1.422H.4.6. Fornell-Larcker
sum_pls_model_mod_med$validity$fl_criteria## PE EE SI HM FC HA IU TRI_A IU*TRI_A TRI_B
## PE 0.815 . . . . . . . . .
## EE 0.493 0.936 . . . . . . . .
## SI 0.561 0.289 0.843 . . . . . . .
## HM 0.641 0.484 0.450 0.920 . . . . . .
## FC 0.542 0.668 0.424 0.570 0.797 . . . . .
## HA 0.551 0.599 0.445 0.639 0.608 0.901 . . . .
## IU 0.650 0.413 0.570 0.616 0.527 0.630 0.910 . . .
## TRI_A 0.600 0.437 0.458 0.596 0.494 0.633 0.589 0.844 . .
## IU*TRI_A -0.017 -0.043 0.166 0.037 0.055 0.098 -0.078 0.136 1.000 .
## TRI_B 0.420 0.617 0.275 0.440 0.583 0.568 0.397 0.429 0.034 0.801
## FC*TRI_B -0.084 -0.223 0.165 -0.130 -0.348 -0.158 -0.069 -0.021 0.259 -0.101
## SNS 0.585 0.528 0.464 0.579 0.544 0.753 0.608 0.526 0.072 0.530
## FC*TRI_B SNS
## PE . .
## EE . .
## SI . .
## HM . .
## FC . .
## HA . .
## IU . .
## TRI_A . .
## IU*TRI_A . .
## TRI_B . .
## FC*TRI_B 1.000 .
## SNS -0.082 0.769
##
## FL Criteria table reports square root of AVE on the diagonal and construct correlations on the lower triangle.write.xlsx2(x=sum_pls_model_mod_med$validity$fl_criteria,
'resumen.xlsx',
sheetName = "Fornell-Larcker_moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)H.4.7. fSquare
sum_pls_model_mod_med$fSquare ## PE EE SI HM FC HA IU TRI_A IU*TRI_A TRI_B
## PE 0.000 0.000 0.000 0.000 0.000 0.000 0.079 0.000 0.000 0.000
## EE 0.000 0.000 0.000 0.000 0.000 0.000 0.008 0.000 0.000 0.000
## SI 0.000 0.000 0.000 0.000 0.000 0.000 0.066 0.000 0.000 0.000
## HM 0.000 0.000 0.000 0.000 0.000 0.000 0.028 0.000 0.000 0.000
## FC 0.000 0.000 0.000 0.000 0.000 0.000 0.007 0.000 0.000 0.000
## HA 0.000 0.000 0.000 0.000 0.000 0.000 0.087 0.000 0.000 0.000
## IU 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_A 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## IU*TRI_A 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_B 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC*TRI_B 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SNS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC*TRI_B SNS
## PE 0.000 0.000
## EE 0.000 0.000
## SI 0.000 0.000
## HM 0.000 0.000
## FC 0.000 .
## HA 0.000 0.285
## IU 0.000 .
## TRI_A 0.000 .
## IU*TRI_A 0.000 0.001
## TRI_B 0.000 .
## FC*TRI_B 0.000 0.004
## SNS 0.000 0.000
##
## The fSquare for certain relationships cannot be calculated as the model contains an interaction term and omitting either the antecedent or moderator in the interaction term will cause model estimation to failwrite.xlsx2(x=sum_pls_model_mod_med$fSquare,
'resumen.xlsx',
sheetName = "fSquare_moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)H.4.8.HTMT
sum_pls_model_mod_med$validity$htmt ## PE EE SI HM FC HA IU TRI_A IU*TRI_A TRI_B
## PE . . . . . . . . . .
## EE 0.556 . . . . . . . . .
## SI 0.657 0.310 . . . . . . . .
## HM 0.737 0.524 0.504 . . . . . . .
## FC 0.695 0.815 0.523 0.707 . . . . . .
## HA 0.624 0.636 0.486 0.692 0.744 . . . . .
## IU 0.795 0.474 0.680 0.724 0.688 0.728 . . . .
## TRI_A 0.704 0.482 0.522 0.669 0.620 0.697 0.707 . . .
## IU*TRI_A 0.078 0.045 0.183 0.039 0.091 0.100 0.087 0.149 . .
## TRI_B 0.508 0.692 0.315 0.496 0.744 0.627 0.481 0.489 0.088 .
## FC*TRI_B 0.092 0.231 0.180 0.137 0.415 0.164 0.077 0.051 0.259 0.108
## SNS 0.730 0.618 0.563 0.689 0.721 0.881 0.777 0.636 0.106 0.655
## FC*TRI_B SNS
## PE . .
## EE . .
## SI . .
## HM . .
## FC . .
## HA . .
## IU . .
## TRI_A . .
## IU*TRI_A . .
## TRI_B . .
## FC*TRI_B . .
## SNS 0.094 .write.xlsx2(x=sum_pls_model_mod_med$validity$htmt ,
'resumen.xlsx',
sheetName = "htmt_moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)H.4.9. Tabla de correlaciones
sum_pls_model_mod_med$descriptives$correlations$constructs ## PE EE SI HM FC HA IU TRI_A IU*TRI_A TRI_B
## PE 1.000 0.493 0.561 0.641 0.542 0.551 0.650 0.600 -0.017 0.420
## EE 0.493 1.000 0.289 0.484 0.668 0.599 0.413 0.437 -0.043 0.617
## SI 0.561 0.289 1.000 0.450 0.424 0.445 0.570 0.458 0.166 0.275
## HM 0.641 0.484 0.450 1.000 0.570 0.639 0.616 0.596 0.037 0.440
## FC 0.542 0.668 0.424 0.570 1.000 0.608 0.527 0.494 0.055 0.583
## HA 0.551 0.599 0.445 0.639 0.608 1.000 0.630 0.633 0.098 0.568
## IU 0.650 0.413 0.570 0.616 0.527 0.630 1.000 0.589 -0.078 0.397
## TRI_A 0.600 0.437 0.458 0.596 0.494 0.633 0.589 1.000 0.136 0.429
## IU*TRI_A -0.017 -0.043 0.166 0.037 0.055 0.098 -0.078 0.136 1.000 0.034
## TRI_B 0.420 0.617 0.275 0.440 0.583 0.568 0.397 0.429 0.034 1.000
## FC*TRI_B -0.084 -0.223 0.165 -0.130 -0.348 -0.158 -0.069 -0.021 0.259 -0.101
## SNS 0.585 0.528 0.464 0.579 0.544 0.753 0.608 0.526 0.072 0.530
## FC*TRI_B SNS
## PE -0.084 0.585
## EE -0.223 0.528
## SI 0.165 0.464
## HM -0.130 0.579
## FC -0.348 0.544
## HA -0.158 0.753
## IU -0.069 0.608
## TRI_A -0.021 0.526
## IU*TRI_A 0.259 0.072
## TRI_B -0.101 0.530
## FC*TRI_B 1.000 -0.082
## SNS -0.082 1.000write.xlsx2(x=sum_pls_model_mod_med$descriptives$correlations$constructs ,
'resumen.xlsx',
sheetName = "Correl_constructos_moderador",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)H.4.10. Otros
b) Efectos totales
c) Efectos indirectos
d) Puntuaciones estimadas para los constructos
e) seleccion de modelo BIC, AIC
sum_pls_model_mod_med$total_effects ## b)## PE EE SI HM FC HA IU TRI_A IU*TRI_A TRI_B
## PE 0.000 0.000 0.000 0.000 0.000 0.000 0.269 0.000 0.000 0.000
## EE 0.000 0.000 0.000 0.000 0.000 0.000 -0.085 0.000 0.000 0.000
## SI 0.000 0.000 0.000 0.000 0.000 0.000 0.209 0.000 0.000 0.000
## HM 0.000 0.000 0.000 0.000 0.000 0.000 0.159 0.000 0.000 0.000
## FC 0.000 0.000 0.000 0.000 0.000 0.000 0.084 0.000 0.000 0.000
## HA 0.000 0.000 0.000 0.000 0.000 0.000 0.286 0.000 0.000 0.000
## IU 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_A 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## IU*TRI_A 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_B 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC*TRI_B 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SNS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC*TRI_B SNS
## PE 0.000 0.055
## EE 0.000 -0.017
## SI 0.000 0.043
## HM 0.000 0.033
## FC 0.000 0.084
## HA 0.000 0.590
## IU 0.000 0.205
## TRI_A 0.000 -0.017
## IU*TRI_A 0.000 0.021
## TRI_B 0.000 0.119
## FC*TRI_B 0.000 0.040
## SNS 0.000 0.000sum_pls_model_mod_med$total_indirect_effects ## c)## PE EE SI HM FC HA IU TRI_A IU*TRI_A TRI_B
## PE 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## EE 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SI 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## HM 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## HA 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## IU 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_A 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## IU*TRI_A 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_B 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC*TRI_B 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## SNS 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## FC*TRI_B SNS
## PE 0.000 0.055
## EE 0.000 -0.017
## SI 0.000 0.043
## HM 0.000 0.033
## FC 0.000 0.017
## HA 0.000 0.059
## IU 0.000 0.000
## TRI_A 0.000 0.000
## IU*TRI_A 0.000 0.000
## TRI_B 0.000 0.000
## FC*TRI_B 0.000 0.000
## SNS 0.000 0.000# sum_pls_model_mod_med$composite_scores ## d)
sum_pls_model_mod_med$it_criteria ## e)## IU SNS
## AIC -320.516 -348.341
## BIC -292.880 -316.757H.5. Evaluar Boot Moderador
sum_boot_pls_model_mod$bootstrapped_paths## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## PE -> IU 0.269 0.268 0.057 4.740 0.148
## EE -> IU -0.085 -0.081 0.048 -1.771 -0.184
## SI -> IU 0.209 0.210 0.053 3.946 0.106
## HM -> IU 0.159 0.156 0.051 3.139 0.045
## FC -> IU 0.084 0.085 0.056 1.499 -0.023
## FC -> SNS 0.066 0.065 0.048 1.396 -0.030
## HA -> IU 0.286 0.287 0.052 5.544 0.189
## HA -> SNS 0.531 0.534 0.053 9.952 0.429
## IU -> SNS 0.205 0.206 0.047 4.369 0.122
## TRI_A -> SNS -0.017 -0.017 0.045 -0.364 -0.111
## IU*TRI_A -> SNS 0.021 0.023 0.042 0.508 -0.054
## TRI_B -> SNS 0.119 0.117 0.045 2.678 0.027
## FC*TRI_B -> SNS 0.040 0.043 0.034 1.179 -0.028
## 97.5% CI
## PE -> IU 0.382
## EE -> IU 0.011
## SI -> IU 0.315
## HM -> IU 0.249
## FC -> IU 0.197
## FC -> SNS 0.156
## HA -> IU 0.384
## HA -> SNS 0.640
## IU -> SNS 0.301
## TRI_A -> SNS 0.066
## IU*TRI_A -> SNS 0.112
## TRI_B -> SNS 0.203
## FC*TRI_B -> SNS 0.103sum_boot$bootstrapped_HTMT ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## PE -> EE 0.556 0.556 0.045 12.385 0.464 0.641
## PE -> SI 0.657 0.660 0.048 13.782 0.568 0.753
## PE -> FC 0.695 0.698 0.052 13.367 0.594 0.790
## PE -> HM 0.737 0.737 0.035 21.122 0.670 0.800
## PE -> HA 0.624 0.625 0.040 15.430 0.540 0.698
## PE -> IU 0.795 0.796 0.038 20.780 0.723 0.864
## PE -> SNS 0.730 0.731 0.048 15.137 0.636 0.817
## EE -> SI 0.310 0.311 0.055 5.681 0.198 0.414
## EE -> FC 0.815 0.815 0.035 23.482 0.740 0.878
## EE -> HM 0.524 0.522 0.045 11.628 0.437 0.600
## EE -> HA 0.636 0.637 0.034 18.793 0.575 0.702
## EE -> IU 0.474 0.473 0.050 9.402 0.367 0.569
## EE -> SNS 0.618 0.617 0.040 15.340 0.534 0.687
## SI -> FC 0.523 0.531 0.064 8.182 0.401 0.645
## SI -> HM 0.504 0.506 0.049 10.219 0.412 0.601
## SI -> HA 0.486 0.484 0.047 10.344 0.385 0.576
## SI -> IU 0.680 0.680 0.048 14.067 0.581 0.774
## SI -> SNS 0.563 0.563 0.050 11.173 0.461 0.661
## FC -> HM 0.707 0.708 0.041 17.348 0.618 0.786
## FC -> HA 0.744 0.746 0.043 17.333 0.659 0.825
## FC -> IU 0.688 0.689 0.043 16.039 0.609 0.784
## FC -> SNS 0.721 0.722 0.041 17.679 0.645 0.794
## HM -> HA 0.692 0.691 0.034 20.184 0.620 0.753
## HM -> IU 0.724 0.724 0.036 20.304 0.653 0.791
## HM -> SNS 0.689 0.689 0.042 16.256 0.607 0.775
## HA -> IU 0.728 0.729 0.041 17.779 0.643 0.805
## HA -> SNS 0.881 0.880 0.023 37.923 0.833 0.921
## IU -> SNS 0.777 0.777 0.042 18.695 0.696 0.858summary_estimacion_model$validity$htmt## PE EE SI FC HM HA IU SNS
## PE . . . . . . . .
## EE 0.556 . . . . . . .
## SI 0.657 0.310 . . . . . .
## FC 0.695 0.815 0.523 . . . . .
## HM 0.737 0.524 0.504 0.707 . . . .
## HA 0.624 0.636 0.486 0.744 0.692 . . .
## IU 0.795 0.474 0.680 0.688 0.724 0.728 . .
## SNS 0.730 0.618 0.563 0.721 0.689 0.881 0.777 .H.6. Simple slope analysis plot
slope_analysis(
moderated_model = pls_model_mod_med,
dv = 'SNS',
moderator = 'TRI_A',
iv = 'IU',
leg_place = 'bottomright')
#plot_interaction(pls_model_mod_med, 'IU*TRI_A', 'SNS')slope_analysis(
moderated_model = pls_model_mod_med,
dv = 'SNS',
moderator = 'TRI_B',
iv = 'FC',
leg_place = 'bottomright')
#plot_interaction(pls_model_mod_med, 'FC*TRI_B', 'SNS')I. Comparación con otros modelos
I.1 Creamos los modelos adicionales.
Nota: No se modificará el modelo de medida. Comparación es a nivel de modelo estructural
#modelo 0 #modelo evaluado creado en E.2
#modelo_estruc <- relationships(
# paths(from = c('PE', 'EE', 'SI', 'FC', 'HM', "HA"), to = c('IU')),
# paths(from = c('FC', 'HA', "IU"), to = c('SNS'))
# )
# Modelo 1
structural_model1 <- relationships(
paths(from = c('PE', 'EE', 'SI', 'FC', 'HM', "HA"), to = c('IU')),
paths(from = c('HA', "IU"), to = c('SNS'))
)
# Modelo 2
structural_model2 <- relationships(
paths(from = c('PE', 'EE', 'SI', 'FC', 'HM', "HA"), to = c('IU')),
paths(from = c("IU"), to = c('SNS'))
)
# Modelo 3
structural_model3 <- relationships(
paths(from = c('PE', 'EE', 'SI', 'HM' ), to = c('IU')),
paths(from = c( 'HA', 'FC','IU'), to = c('SNS'))
)plot(modelo_estruc) # Modelo inicial plot(structural_model1)plot(structural_model2)plot(structural_model3)I.2 Generamos los modelos
pls_model1 <- estimate_pls(
data = pls_data2,
measurement_model = modelo_medida,
structural_model = structural_model1,
missing_value = '-99'
)
sum_model1 <- summary(pls_model1)
pls_model2 <- estimate_pls(
data = pls_data2,
measurement_model = modelo_medida,
structural_model = structural_model2,
missing_value = '-99'
)
sum_model2 <- summary(pls_model2)
pls_model3 <- estimate_pls(
data = pls_data2,
measurement_model = modelo_medida,
structural_model = structural_model3,
missing_value = '-99'
)
sum_model3 <- summary(pls_model3)I.3 Comparamos los modelos
summary_estimacion_model$it_criteria## IU SNS
## AIC -320.517 -345.750
## BIC -292.881 -329.958sum_model1$it_criteria## IU SNS
## AIC -320.542 -343.473
## BIC -292.906 -331.629sum_model2$it_criteria## IU SNS
## AIC -320.758 -177.370
## BIC -293.122 -169.474sum_model3$it_criteria## IU SNS
## AIC -286.121 -346.233
## BIC -266.381 -330.441# Menor BIC tiene mejor poder predictivo# Recogemos los valores BIC de cada modelo.
#Nos centramos en este ya que es el que intermedia, el que esta cambiando los modelos
itcriteria_vector <- c(summary_estimacion_model$it_criteria['BIC', 'IU'],
sum_model1$it_criteria['BIC', 'IU'],
sum_model2$it_criteria['BIC', 'IU'],
sum_model3$it_criteria['BIC', 'IU'])
itcriteria_vector2 <- c(summary_estimacion_model$it_criteria['BIC', 'SNS'],
sum_model1$it_criteria['BIC', 'SNS'],
sum_model2$it_criteria['BIC', 'SNS'],
sum_model3$it_criteria['BIC', 'SNS'])
# Assign the model names to IT Criteria vector
names(itcriteria_vector) <- c('Original','Model1', 'Model2', 'Model3')
names(itcriteria_vector2) <- c('Original','Model1', 'Model2', 'Model3')# Valores BIC por modelos # El menor BIC seleccionamos - IU
itcriteria_vector## Original Model1 Model2 Model3
## -292.8812 -292.9058 -293.1221 -266.3810# Calcula BIC Akaike # Mayor implica mejor poder predictivo - IU
compute_itcriteria_weights(itcriteria_vector)## Original Model1 Model2 Model3
## 3.184311e-01 3.223746e-01 3.591937e-01 5.605111e-07# Valores BIC para SNS en distintos modelos - SNS
itcriteria_vector2## Original Model1 Model2 Model3
## -329.9579 -331.6285 -169.4738 -330.4410# Calcula BIC Akaike # Mayor implica mejor poder predictivo -SNS
compute_itcriteria_weights(itcriteria_vector2)## Original Model1 Model2 Model3
## 2.183965e-01 5.035313e-01 3.094420e-36 2.780722e-01J. Análisis Multigrupo
Asumiremos que se desea crear multigrupo con la variable género.
NOTA: Solo se puede hacer multigrupo de 2 grupos. Más grupos no es posible en esta versión.
NOTA2: Cambiaremos el modelo estructural para que MGA sea significativo
modelo_estruc_mga <- relationships(
paths(from = c('PE', 'SI', "HA"), to = c('IU')),
paths(from = c('HA', "IU"), to = c('SNS'))
)
plot(modelo_estruc_mga)mga_esti <- estimate_pls(data = pls_data2,
measurement_model = modelo_medida, #E1
structural_model = modelo_estruc_mga,
missing = mean_replacement,
missing_value = -99)J.1. Preparación de variable
En caso que no se haya convertido en D.2
#pls_data2$GENDER
#pls_data2$GENERO = ifelse(pls_data2$GENDER=='Male', 1, 2)
#pls_data2$GENDER
#pls_data2$REGION3= ifelse(pls_data2$REGION=='Coquimbo', 1, 2) #59sum(pls_data2$GENERO==1) #Male## [1] 170sum(pls_data2$GENERO==2)## [1] 213J.2. Generamos el multigrupo
En este caso probaremos 2 MGA uno con el Género y otro con la región
pls_mga <- estimate_pls_mga(mga_esti,
pls_data2$GENERO == 1,
nboot=500) ## sobre 2000pls_mga_region <- estimate_pls_mga(mga_esti,
pls_data2$REGION3 == 1,
nboot=500) ## sobre 2000J.3. Análisis del Multigrupo
Desde <- pls_mga$source
Hasta <- pls_mga$target
Grupo_1 <- pls_mga$group1_beta
Grupo_2 <- pls_mga$group2_beta
p_value <- pls_mga$pls_mga_p
mga_1 <- data.frame(Desde, Hasta, "B Grupo1" = Grupo_1, "B Grupo2" = Grupo_2, p_value)
mga_1# p-values <0.05 significa que hay diferencia significativa, entre los grupos por cada Hipo.Desde <- pls_mga_region$source
Hasta <- pls_mga_region$target
Grupo_1 <- pls_mga_region$group1_beta
Grupo_2 <- pls_mga_region$group2_beta
p_value <- pls_mga_region$pls_mga_p
mga_2 <- data.frame(Desde, Hasta, "B Grupo1" = Grupo_1, "B Grupo2" = Grupo_2, p_value)
mga_2# p-values <0.05 significa que hay diferencia significativa, entre los grupos por cada Hipo.write.xlsx2(x=pls_mga ,
'resumen.xlsx',
sheetName = "MGA",
col.names = TRUE,
row.names = TRUE,
append = TRUE,
showNA = TRUE,
password = NULL)J.4. Análisis MICOM
NOTA: Utilizaremos paquete cSEM y sentencia en Lavaan
Modelo de medida y estructural
cSmodel2 <- "
# modelo estructural
SNS ~ IU + HA
IU ~ HA + SI + PE
# modelo de medida
PE =~ PE1 + PE2 + PE3 + PE4
SI =~ SI1 + SI2 + SI3 + SI4
HA =~ HA1 + HA2 + HA3 + HA4 + HA5
IU =~ IU1 + IU2
SNS =~ U1 + U2+ U3 + U4
"Generamos data y probamos los modelos
#1 Data Genero
g11 <- pls_data2[(pls_data2$GENERO==1),]
g12 <- pls_data2[(pls_data2$GENERO!=1 ),]
#2 Data región
g21 <- pls_data2[(pls_data2$REGION=='Coquimbo'),]
g22 <- pls_data2[(pls_data2$REGION!='Coquimbo' ),]
csem_results1 <- csem(.data = g11, cSmodel2)
csem_results2 <- csem(.data = g12, cSmodel2)
## Analisis con cSEM
csem_results1 <- csem(.data = g11, cSmodel2)
csem_results2 <- csem(.data = g12, cSmodel2)
#Si en Status da "not Ok", no se puede usar para MGA
verify(csem_results1)## ________________________________________________________________________________
##
## Verify admissibility:
##
## admissible
##
## Details:
##
## Code Status Description
## 1 ok Convergence achieved
## 2 ok All absolute standardized loading estimates <= 1
## 3 ok Construct VCV is positive semi-definite
## 4 ok All reliability estimates <= 1
## 5 ok Model-implied indicator VCV is positive semi-definite
## ________________________________________________________________________________verify(csem_results2)## ________________________________________________________________________________
##
## Verify admissibility:
##
## admissible
##
## Details:
##
## Code Status Description
## 1 ok Convergence achieved
## 2 ok All absolute standardized loading estimates <= 1
## 3 ok Construct VCV is positive semi-definite
## 4 ok All reliability estimates <= 1
## 5 ok Model-implied indicator VCV is positive semi-definite
## ________________________________________________________________________________## Analisis con cSEM
csem_results1 <- csem(.data = g21, cSmodel2)
csem_results2 <- csem(.data = g22, cSmodel2)
#Si en Status da "not Ok", no se puede usar para MGA
verify(csem_results1)## ________________________________________________________________________________
##
## Verify admissibility:
##
## inadmissible
##
## Details:
##
## Code Status Description
## 1 ok Convergence achieved
## 2 not ok All absolute standardized loading estimates <= 1
## 3 ok Construct VCV is positive semi-definite
## 4 ok All reliability estimates <= 1
## 5 ok Model-implied indicator VCV is positive semi-definite
## ________________________________________________________________________________verify(csem_results2)## ________________________________________________________________________________
##
## Verify admissibility:
##
## admissible
##
## Details:
##
## Code Status Description
## 1 ok Convergence achieved
## 2 ok All absolute standardized loading estimates <= 1
## 3 ok Construct VCV is positive semi-definite
## 4 ok All reliability estimates <= 1
## 5 ok Model-implied indicator VCV is positive semi-definite
## ________________________________________________________________________________Test MICOM
csem_results <- csem(.data = list("group1" = g11, "group2" = g12), # Data creada por grupo
cSmodel2, .resample_method = "bootstrap",
.R = 500) ##Subir numero
testMICOM(csem_results,
.R = 500) ##Subir numero## ________ Test for measurement invariance based on Henseler et al (2016) ________## ________________________________________________________________________________
## -------- Test for measurement invariance based on Henseler et al (2016) --------
## ======================== Step 1 - Configural invariance ========================
##
## Configural invariance is a precondition for step 2 and 3.
## Do not proceed to interpret results unless
## configural invariance has been established.
##
## ======================= Step 2 - Compositional invariance ======================
##
## Null hypothesis:
##
## +-----------------------------------------------------------------+
## | |
## | H0: Compositional measurement invariance of the constructs. |
## | |
## +-----------------------------------------------------------------+
##
## Test statistic and p-value:
##
## Compared groups: group1_group2
## p-value by adjustment
## Construct Test statistic none
## HA 1.0000 0.9345
## SI 0.9980 0.2326
## PE 0.9995 0.8436
## IU 1.0000 0.6596
## SNS 0.9995 0.5793
##
##
## ================= Step 3 - Equality of the means and variances =================
##
## Null hypothesis:
##
## +------------------------------------------------------------+
## | |
## | 1. H0: Difference between group means is zero |
## | 2. H0: Log of the ratio of the group variances is zero |
## | |
## +------------------------------------------------------------+
##
## Test statistic and critical values:
##
## Compared groups: group1_group2
##
## Mean
## p-value by adjustment
## Construct Test statistic none
## HA 0.0153 0.8740
## SI -0.0230 0.8300
## PE 0.1527 0.1520
## IU -0.0701 0.5220
## SNS 0.0132 0.9160
##
## Var
## p-value by adjustment
## Construct Test statistic none
## HA -0.1683 0.1200
## SI -0.0284 0.8280
## PE -0.2563 0.1040
## IU -0.1839 0.3620
## SNS -0.0355 0.7780
##
##
## Additional information:
##
## Out of 500 permutation runs, 473 where admissible.
## See ?verify() for what constitutes an inadmissible result.
##
## The seed used was: -1395106298
##
## Number of observations per group:
##
## Group No. observations
## group1 170
## group2 213
## ________________________________________________________________________________Test de comparacion MGA
testmgd <- testMGD(csem_results, .parameters_to_compare = NULL,
.alpha = 0.05,
.approach_p_adjust = c("none", "bonferroni"), ## Tipo de ajuste a los p
.R_permutation = 60,
.R_bootstrap = 60, #Subir numero
.saturated = FALSE,
.approach_mgd = "all", #test a aplicar
.output_type = "complete", #"c("complete", "structured"),
.eval_plan = c("sequential", "multicore", "multisession"),
.verbose = FALSE)## Warning: The following warning occured in the testMGD() function:
## Currently, there is no p-value adjustment possible for the approach suggested by
## Henseler (2007), CI_para, and CI_overlap. Adjustment is ignored for these approaches.###Test no rechazarán sus respectivas H0, los grupos son prácticamente idénticos.
testmgd## ________________________________________________________________________________
## ----------------------------------- Overview -----------------------------------
##
## Total permutation runs = 62
## Admissible permutation results = 60
## Permutation seed = -1257783269
##
## Total bootstrap runs = 500
## Admissible bootstrap results:
##
## Group Admissibles
## group1 371
## group2 483
##
## Bootstrap seed:
##
## Group Seed
## group1 317975023
## group2 -640193739
##
## Number of observations per group:
##
## Group No. Obs.
## group1 170
## group2 213
##
## Overall decision (based on alpha = 5%):
##
## p_adjust = 'none'p_adjust = 'bonferroni'
## Sarstedt reject reject
## Chin Do not reject Do not reject
## Keil Do not reject Do not reject
## Nitzl Do not reject Do not reject
##
## For details on a particular approach type:
##
## - `print(<object-name>, .approach_mgd = '<approach-name>')`
##
## ________________________________________________________________________________K. Análisis Segundo Orden
Data contiene TRI el cual está conformado por 4 constructos, asumiremos que corresponde a un constructo de segundo orden el qye afecta a IU
K.1. Evaluar constructos de orden inferior
K.1.1. Modelo de medida
K.1.1.a Modelo de medida Formativo
m_medida_1 <- constructs(
composite('TRI_A', multi_items('TRI', 1:4), weights = mode_B ), #Formativo de ejemplo
composite('TRI_B', multi_items('TRI', 5:8), weights = mode_B ),
composite('TRI_C', multi_items('TRI', 9:12), weights = mode_B ),
composite('TRI_D', multi_items('TRI', 13:16), weights = mode_B ),
composite('IU', multi_items('IU', 1:2)),
composite('SNS', multi_items('U', 1:4))
)
plot(m_medida_1)K.1.2. Modelo de medida Reflectivo
m_medida_2 <- constructs(
composite('TRI_A', multi_items('TRI', 1:4)),
composite('TRI_B', multi_items('TRI', 5:8)),
composite('TRI_C', multi_items('TRI', 9:12)),
composite('TRI_D', multi_items('TRI', 13:16) ),
composite('IU', multi_items('IU', 1:2)),
composite('SNS', multi_items('U', 1:4))
)
plot(m_medida_2)K.1.2. Modelo estructural
m_estruc_1 <- relationships(
paths(from = c('TRI_A', 'TRI_B', 'TRI_C', 'TRI_D'), to = c('IU')),
paths(from = c("IU"), to = c('SNS'))
)
plot(m_estruc_1)K.1.3. Estimación modelo
estimacion_model_1 <- estimate_pls(data = pls_data2,
measurement_model = m_medida_1, #K.1.1. - modelo de medida
structural_model = m_estruc_1, #K.1.2. - modelo estructural
inner_weights = path_weighting,
# path_weighting para path weighting (default) o path_factorial para factor weighting,
missing = mean_replacement,
missing_value = '-99' )
summary_m_1 = summary(estimacion_model_1)
plot(estimacion_model_1)K.1.4. Evaluación del modelo de orden inferior
plot(summary_m_1$reliability, title = "Fig. : Fiabilidad orden inferior")
plot(summary_m_1$paths[,1], pch = 2, col = "red", main="Betas y R^2 (Exogenos)",
xlab = "Variables", ylab = "Valores estimados", xlim = c(0,length(row.names(summary_m_1$paths))+1)
)
text(summary_m_1$paths[,1],labels = row.names(summary_m_1$paths) , pos = 4)
plot(summary_m_1$paths[,2], pch = 2, col = "red", main="Betas y R^2 (Endogenos)",
xlab = "Variables", ylab = "Valores estimados" , xlim = c(0,length(row.names(summary_m_1$paths))+1) )
text(summary_m_1$paths[,2],labels = row.names(summary_m_1$paths) , pos = 4)
summary_m_1$reliability## alpha rhoC AVE rhoA
## TRI_A 0.866 0.895 0.683 1.000
## TRI_B 0.816 0.841 0.577 1.000
## TRI_C 0.797 0.786 0.497 1.000
## TRI_D 0.695 0.532 0.333 1.000
## IU 0.794 0.906 0.829 0.794
## SNS 0.769 0.852 0.591 0.778
##
## Alpha, rhoC, and rhoA should exceed 0.7 while AVE should exceed 0.5summary_m_1$loading## TRI_A TRI_B TRI_C TRI_D IU SNS
## TRI1 0.718 0.000 -0.000 -0.000 0.000 0.000
## TRI2 0.794 0.000 -0.000 -0.000 0.000 0.000
## TRI3 0.858 0.000 -0.000 -0.000 0.000 0.000
## TRI4 0.923 0.000 -0.000 -0.000 0.000 0.000
## TRI5 0.000 0.792 -0.000 -0.000 0.000 0.000
## TRI6 0.000 0.562 -0.000 -0.000 0.000 0.000
## TRI7 0.000 0.926 -0.000 -0.000 0.000 0.000
## TRI8 0.000 0.713 -0.000 -0.000 0.000 0.000
## TRI9 -0.000 -0.000 0.736 0.000 -0.000 -0.000
## TRI10 -0.000 -0.000 0.955 0.000 -0.000 -0.000
## TRI11 -0.000 -0.000 0.449 0.000 -0.000 -0.000
## TRI12 -0.000 -0.000 0.576 0.000 -0.000 -0.000
## TRI13 0.000 0.000 0.000 -0.157 0.000 0.000
## TRI14 -0.000 -0.000 0.000 0.499 -0.000 -0.000
## TRI15 -0.000 -0.000 0.000 0.494 -0.000 -0.000
## TRI16 -0.000 -0.000 0.000 0.904 -0.000 -0.000
## IU1 0.000 0.000 -0.000 -0.000 0.907 0.000
## IU2 0.000 0.000 -0.000 -0.000 0.913 0.000
## U1 0.000 0.000 -0.000 -0.000 0.000 0.763
## U2 0.000 0.000 -0.000 -0.000 0.000 0.793
## U3 0.000 0.000 -0.000 -0.000 0.000 0.814
## U4 0.000 0.000 -0.000 -0.000 0.000 0.700summary_m_1$validity$fl_criteria ## TRI_A TRI_B TRI_C TRI_D IU SNS
## TRI_A 0.827 . . . . .
## TRI_B 0.475 0.760 . . . .
## TRI_C -0.335 -0.471 0.705 . . .
## TRI_D -0.441 -0.483 0.468 0.577 . .
## IU 0.602 0.415 -0.347 -0.388 0.910 .
## SNS 0.531 0.547 -0.372 -0.453 0.612 0.769
##
## FL Criteria table reports square root of AVE on the diagonal and construct correlations on the lower triangle.summary_m_1$validity$htmt ## TRI_A TRI_B TRI_C TRI_D IU SNS
## TRI_A . . . . . .
## TRI_B 0.489 . . . . .
## TRI_C 0.303 0.610 . . . .
## TRI_D 0.406 0.550 0.558 . . .
## IU 0.707 0.481 0.375 0.370 . .
## SNS 0.636 0.655 0.460 0.485 0.777 .summary_m_1$validity$vif_items ## TRI_A :
## TRI1 TRI2 TRI3 TRI4
## 1.967 2.076 2.563 2.065
##
## TRI_B :
## TRI5 TRI6 TRI7 TRI8
## 1.785 1.746 1.930 1.636
##
## TRI_C :
## TRI9 TRI10 TRI11 TRI12
## 1.402 1.733 1.767 2.054
##
## TRI_D :
## TRI13 TRI14 TRI15 TRI16
## 1.239 1.663 1.690 1.375
##
## IU :
## IU1 IU2
## 1.764 1.764
##
## SNS :
## U1 U2 U3 U4
## 1.459 1.501 1.708 1.422K.2. Constructo de orden superior
K.2.1. Modelo de medida
a. Modelo de medida Formativo
m_medida_3 <- constructs(
composite('TRI_A', multi_items('TRI', 1:4), weights = mode_B),
composite('TRI_B', multi_items('TRI', 5:8), weights = mode_B),
composite('TRI_C', multi_items('TRI', 9:12), weights = mode_B),
composite('TRI_D', multi_items('TRI', 13:16), weights = mode_B),
higher_composite('TRI', c('TRI_A', 'TRI_B', 'TRI_C', 'TRI_D'), method ='two stage', weights = mode_B),
composite('IU', multi_items('IU', 1:2)),
composite('SNS', multi_items('U', 1:4))
)
plot(m_medida_3) b. Modelo de medida Reflectivo
m_medida_4 <- constructs(
composite('TRI_A', multi_items('TRI', 1:4)),
composite('TRI_B', multi_items('TRI', 5:8)),
composite('TRI_C', multi_items('TRI', 9:12)),
composite('TRI_D', multi_items('TRI', 13:16)),
higher_composite('TRI', c('TRI_A', 'TRI_B', 'TRI_C', 'TRI_D'), method ='two stage', weights = mode_B),
composite('IU', multi_items('IU', 1:2)),
composite('SNS', multi_items('U', 1:4))
)
plot(m_medida_4) K.2.2. Modelo estructural
m_estruc_2 <- relationships(
paths(from = 'TRI', to = 'IU'),
paths(from = c("IU"), to = c('SNS')))
plot(m_estruc_2)K.2.3. Estimación modelo
a. Estimación modelo Formativo
estimacion_model_2 <- estimate_pls(data = pls_data2,
measurement_model = m_medida_3, #K.2.1. a
structural_model = m_estruc_2, # K.2.2.
inner_weights = path_weighting,
# path_weighting para path weighting (default) o path_factorial para factor weighting,
missing = mean_replacement, #Reemplazar los valores perdido mean es default
missing_value = '-99' )
plot(estimacion_model_2)summary_m_2 = summary(estimacion_model_2)b. Estimación modelo Reflectivo
estimacion_model_3 <- estimate_pls(data = pls_data2,
measurement_model = m_medida_4, #K.2.1. b
structural_model = m_estruc_2, # K.2.2.
inner_weights = path_weighting,
# path_weighting para path weighting (default) o path_factorial para factor weighting,
missing = mean_replacement, #Reemplazar los valores perdido mean es default
missing_value = '-99' )
plot(estimacion_model_3)summary_m_3 = summary(estimacion_model_3)K.2.4. Evaluación modelo de 2do orden Formativo
plot(summary_m_2$reliability, title = "Fig. : Fiabilidad orden inferior")
plot(summary_m_2$paths[,1], pch = 2, col = "red", main="Betas y R^2 (Exogenos)",
xlab = "Variables", ylab = "Valores estimados", xlim = c(0,length(row.names(summary_m_2$paths))+1)
)
text(summary_m_2$paths[,1],labels = row.names(summary_m_2$paths) , pos = 4)
plot(summary_m_2$paths[,2], pch = 2, col = "red", main="Betas y R^2 (Endogenos)",
xlab = "Variables", ylab = "Valores estimados" , xlim = c(0,length(row.names(summary_m_2$paths))+1) )
text(summary_m_2$paths[,2],labels = row.names(summary_m_2$paths) , pos = 4)
summary_m_2$reliability## alpha rhoC AVE rhoA
## TRI -0.864 0.092 0.504 1.000
## IU 0.794 0.906 0.829 0.794
## SNS 0.769 0.852 0.591 0.778
## TRI_A 0.866 0.895 0.683 1.000
## TRI_B 0.816 0.841 0.577 1.000
## TRI_C 0.797 0.786 0.497 1.000
## TRI_D 0.695 0.532 0.333 1.000
##
## Alpha, rhoC, and rhoA should exceed 0.7 while AVE should exceed 0.5summary_m_2$loading## TRI IU SNS TRI_A TRI_B TRI_C TRI_D
## TRI_A 0.953 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_B 0.657 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_C -0.549 -0.000 -0.000 0.000 0.000 0.000 0.000
## TRI_D -0.613 -0.000 -0.000 0.000 0.000 0.000 0.000
## IU1 0.000 0.907 0.000 0.000 0.000 0.000 0.000
## IU2 0.000 0.913 0.000 0.000 0.000 0.000 0.000
## U1 0.000 0.000 0.763 0.000 0.000 0.000 0.000
## U2 0.000 0.000 0.793 0.000 0.000 0.000 0.000
## U3 0.000 0.000 0.814 0.000 0.000 0.000 0.000
## U4 0.000 0.000 0.700 0.000 0.000 0.000 0.000
## TRI1 0.000 0.000 0.000 0.718 0.000 -0.000 -0.000
## TRI2 0.000 0.000 0.000 0.794 0.000 -0.000 -0.000
## TRI3 0.000 0.000 0.000 0.858 0.000 -0.000 -0.000
## TRI4 0.000 0.000 0.000 0.923 0.000 -0.000 -0.000
## TRI5 0.000 0.000 0.000 0.000 0.792 -0.000 -0.000
## TRI6 0.000 0.000 0.000 0.000 0.562 -0.000 -0.000
## TRI7 0.000 0.000 0.000 0.000 0.926 -0.000 -0.000
## TRI8 0.000 0.000 0.000 0.000 0.713 -0.000 -0.000
## TRI9 0.000 0.000 0.000 -0.000 -0.000 0.736 0.000
## TRI10 0.000 0.000 0.000 -0.000 -0.000 0.955 0.000
## TRI11 0.000 0.000 0.000 -0.000 -0.000 0.449 0.000
## TRI12 0.000 0.000 0.000 -0.000 -0.000 0.576 0.000
## TRI13 0.000 0.000 0.000 0.000 0.000 0.000 -0.157
## TRI14 0.000 0.000 0.000 -0.000 -0.000 0.000 0.499
## TRI15 0.000 0.000 0.000 -0.000 -0.000 0.000 0.494
## TRI16 0.000 0.000 0.000 -0.000 -0.000 0.000 0.904summary_m_2$validity$fl_criteria ## TRI IU SNS TRI_A TRI_B TRI_C TRI_D
## TRI 0.710 . . . . . .
## IU 0.632 0.910 . . . . .
## SNS 0.610 0.612 0.769 . . . .
## TRI_A 0.953 0.602 0.531 0.827 . . .
## TRI_B 0.657 0.415 0.547 0.475 0.760 . .
## TRI_C -0.549 -0.347 -0.372 -0.335 -0.471 0.705 .
## TRI_D -0.613 -0.388 -0.453 -0.441 -0.483 0.468 0.577
##
## FL Criteria table reports square root of AVE on the diagonal and construct correlations on the lower triangle.summary_m_2$validity$htmt ## TRI IU SNS TRI_A TRI_B TRI_C TRI_D
## TRI . . . . . . .
## IU 0.736 . . . . . .
## SNS 0.816 0.777 . . . . .
## TRI_A 0.870 0.707 0.636 . . . .
## TRI_B 0.943 0.481 0.655 0.489 . . .
## TRI_C 0.899 0.375 0.460 0.303 0.610 . .
## TRI_D 0.796 0.370 0.485 0.406 0.550 0.558 .summary_m_2$validity$vif_items ## TRI :
## TRI_A TRI_B TRI_C TRI_D
## 1.402 1.593 1.429 1.536
##
## IU :
## IU1 IU2
## 1.764 1.764
##
## SNS :
## U1 U2 U3 U4
## 1.459 1.501 1.708 1.422
##
## TRI_A :
## TRI1 TRI2 TRI3 TRI4
## 1.967 2.076 2.563 2.065
##
## TRI_B :
## TRI5 TRI6 TRI7 TRI8
## 1.785 1.746 1.930 1.636
##
## TRI_C :
## TRI9 TRI10 TRI11 TRI12
## 1.402 1.733 1.767 2.054
##
## TRI_D :
## TRI13 TRI14 TRI15 TRI16
## 1.239 1.663 1.690 1.375summary_m_2$validity$cross_loadings## TRI IU SNS TRI_A TRI_B TRI_C TRI_D
## TRI1 0.665 0.432 0.373 0.718 0.304 -0.190 -0.282
## TRI2 0.758 0.478 0.462 0.794 0.355 -0.298 -0.354
## TRI3 0.813 0.517 0.467 0.858 0.357 -0.320 -0.363
## TRI4 0.883 0.555 0.467 0.923 0.476 -0.288 -0.417
## TRI5 0.474 0.329 0.435 0.339 0.792 -0.309 -0.319
## TRI6 0.361 0.234 0.282 0.254 0.562 -0.247 -0.302
## TRI7 0.624 0.385 0.483 0.449 0.926 -0.457 -0.480
## TRI8 0.511 0.296 0.441 0.375 0.713 -0.393 -0.389
## TRI9 -0.366 -0.255 -0.306 -0.168 -0.425 0.736 0.426
## TRI10 -0.542 -0.331 -0.344 -0.351 -0.430 0.955 0.418
## TRI11 -0.271 -0.156 -0.212 -0.125 -0.413 0.449 0.314
## TRI12 -0.356 -0.200 -0.267 -0.203 -0.374 0.576 0.398
## TRI13 0.205 0.061 0.021 0.244 0.014 0.028 -0.157
## TRI14 -0.207 -0.193 -0.228 -0.106 -0.229 0.176 0.499
## TRI15 -0.339 -0.191 -0.312 -0.220 -0.368 0.319 0.494
## TRI16 -0.553 -0.350 -0.441 -0.369 -0.483 0.510 0.904
## TRI_A 0.953 0.602 0.531 1.000 0.475 -0.335 -0.441
## TRI_B 0.657 0.415 0.547 0.475 1.000 -0.471 -0.483
## TRI_C -0.549 -0.347 -0.372 -0.335 -0.471 1.000 0.468
## TRI_D -0.613 -0.388 -0.453 -0.441 -0.483 0.468 1.000
## IU1 0.538 0.907 0.578 0.505 0.367 -0.310 -0.335
## IU2 0.612 0.913 0.537 0.589 0.389 -0.321 -0.371
## U1 0.463 0.482 0.763 0.417 0.381 -0.248 -0.343
## U2 0.468 0.522 0.793 0.400 0.444 -0.315 -0.329
## U3 0.533 0.482 0.814 0.494 0.393 -0.300 -0.345
## U4 0.407 0.381 0.700 0.308 0.482 -0.282 -0.395K.2.5. Bootstrap modelo de 2do orden Formativo
boot_m_2 <- bootstrap_model(seminr_model = estimacion_model_2 , #K.2.3. a
nboot = 500, ### N° Subsamples 5000<
cores = parallel::detectCores(), #CPU cores -parallel processing
seed = 123)
plot(boot_m_2)sum_boot_m_2 <- summary(boot_m_2, alpha=0.05 ) ### Intervalo de confianza, en este caso es dos colas 90%K.2.6. Evaluación Bootstrap modelo de 2do orden Formativo
sum_boot_m_2$bootstrapped_weights ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## TRI_A -> TRI 0.770 0.758 0.060 12.829 0.631
## TRI_B -> TRI 0.158 0.166 0.071 2.231 0.026
## TRI_C -> TRI -0.159 -0.163 0.071 -2.223 -0.296
## TRI_D -> TRI -0.123 -0.122 0.080 -1.549 -0.281
## IU1 -> IU 0.540 0.541 0.014 39.603 0.515
## IU2 -> IU 0.558 0.558 0.013 41.846 0.534
## U1 -> SNS 0.335 0.334 0.021 15.915 0.293
## U2 -> SNS 0.362 0.362 0.023 15.697 0.320
## U3 -> SNS 0.334 0.334 0.024 14.196 0.289
## U4 -> SNS 0.264 0.265 0.019 13.847 0.226
## TRI1 -> TRI_A 0.093 0.092 0.128 0.723 -0.148
## TRI2 -> TRI_A 0.277 0.276 0.119 2.322 0.036
## TRI3 -> TRI_A 0.232 0.244 0.119 1.949 0.000
## TRI4 -> TRI_A 0.558 0.536 0.112 4.970 0.324
## TRI5 -> TRI_B 0.423 0.412 0.142 2.971 0.103
## TRI6 -> TRI_B -0.113 -0.115 0.134 -0.843 -0.352
## TRI7 -> TRI_B 0.621 0.606 0.142 4.364 0.322
## TRI8 -> TRI_B 0.216 0.221 0.168 1.284 -0.096
## TRI9 -> TRI_C 0.348 0.323 0.166 2.100 0.001
## TRI10 -> TRI_C 0.800 0.792 0.151 5.316 0.461
## TRI11 -> TRI_C -0.064 -0.075 0.182 -0.351 -0.375
## TRI12 -> TRI_C 0.014 0.018 0.185 0.076 -0.344
## TRI13 -> TRI_D -0.419 -0.423 0.125 -3.355 -0.658
## TRI14 -> TRI_D 0.425 0.426 0.131 3.248 0.182
## TRI15 -> TRI_D -0.065 -0.059 0.131 -0.495 -0.307
## TRI16 -> TRI_D 0.835 0.806 0.102 8.170 0.588
## 97.5% CI
## TRI_A -> TRI 0.858
## TRI_B -> TRI 0.299
## TRI_C -> TRI -0.010
## TRI_D -> TRI 0.033
## IU1 -> IU 0.568
## IU2 -> IU 0.585
## U1 -> SNS 0.374
## U2 -> SNS 0.410
## U3 -> SNS 0.383
## U4 -> SNS 0.299
## TRI1 -> TRI_A 0.344
## TRI2 -> TRI_A 0.487
## TRI3 -> TRI_A 0.468
## TRI4 -> TRI_A 0.761
## TRI5 -> TRI_B 0.684
## TRI6 -> TRI_B 0.133
## TRI7 -> TRI_B 0.851
## TRI8 -> TRI_B 0.525
## TRI9 -> TRI_C 0.671
## TRI10 -> TRI_C 1.048
## TRI11 -> TRI_C 0.315
## TRI12 -> TRI_C 0.355
## TRI13 -> TRI_D -0.166
## TRI14 -> TRI_D 0.662
## TRI15 -> TRI_D 0.187
## TRI16 -> TRI_D 0.966summary_m_2$validity$vif_items ## TRI :
## TRI_A TRI_B TRI_C TRI_D
## 1.402 1.593 1.429 1.536
##
## IU :
## IU1 IU2
## 1.764 1.764
##
## SNS :
## U1 U2 U3 U4
## 1.459 1.501 1.708 1.422
##
## TRI_A :
## TRI1 TRI2 TRI3 TRI4
## 1.967 2.076 2.563 2.065
##
## TRI_B :
## TRI5 TRI6 TRI7 TRI8
## 1.785 1.746 1.930 1.636
##
## TRI_C :
## TRI9 TRI10 TRI11 TRI12
## 1.402 1.733 1.767 2.054
##
## TRI_D :
## TRI13 TRI14 TRI15 TRI16
## 1.239 1.663 1.690 1.375sum_boot_m_2$bootstrapped_loadings ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## TRI_A -> TRI 0.953 0.945 0.022 42.457 0.893
## TRI_B -> TRI 0.657 0.653 0.055 11.912 0.548
## TRI_C -> TRI -0.549 -0.554 0.065 -8.490 -0.672
## TRI_D -> TRI -0.613 -0.611 0.059 -10.340 -0.722
## IU1 -> IU 0.907 0.907 0.010 94.655 0.887
## IU2 -> IU 0.913 0.913 0.009 98.926 0.894
## U1 -> SNS 0.763 0.764 0.025 30.934 0.715
## U2 -> SNS 0.793 0.793 0.021 37.472 0.751
## U3 -> SNS 0.814 0.813 0.023 35.072 0.765
## U4 -> SNS 0.700 0.701 0.031 22.364 0.636
## TRI1 -> TRI_A 0.718 0.711 0.070 10.267 0.572
## TRI2 -> TRI_A 0.794 0.789 0.056 14.077 0.671
## TRI3 -> TRI_A 0.858 0.853 0.044 19.554 0.761
## TRI4 -> TRI_A 0.923 0.909 0.036 25.666 0.836
## TRI5 -> TRI_B 0.792 0.774 0.072 11.030 0.607
## TRI6 -> TRI_B 0.562 0.551 0.093 6.026 0.363
## TRI7 -> TRI_B 0.926 0.907 0.047 19.560 0.794
## TRI8 -> TRI_B 0.713 0.705 0.097 7.377 0.481
## TRI9 -> TRI_C 0.736 0.703 0.104 7.091 0.487
## TRI10 -> TRI_C 0.955 0.930 0.051 18.890 0.799
## TRI11 -> TRI_C 0.449 0.428 0.137 3.274 0.153
## TRI12 -> TRI_C 0.576 0.557 0.117 4.909 0.316
## TRI13 -> TRI_D -0.157 -0.165 0.141 -1.112 -0.417
## TRI14 -> TRI_D 0.499 0.491 0.114 4.387 0.251
## TRI15 -> TRI_D 0.494 0.483 0.103 4.779 0.258
## TRI16 -> TRI_D 0.904 0.880 0.054 16.591 0.756
## 97.5% CI
## TRI_A -> TRI 0.978
## TRI_B -> TRI 0.759
## TRI_C -> TRI -0.422
## TRI_D -> TRI -0.492
## IU1 -> IU 0.924
## IU2 -> IU 0.930
## U1 -> SNS 0.808
## U2 -> SNS 0.833
## U3 -> SNS 0.854
## U4 -> SNS 0.756
## TRI1 -> TRI_A 0.833
## TRI2 -> TRI_A 0.882
## TRI3 -> TRI_A 0.932
## TRI4 -> TRI_A 0.973
## TRI5 -> TRI_B 0.895
## TRI6 -> TRI_B 0.730
## TRI7 -> TRI_B 0.979
## TRI8 -> TRI_B 0.859
## TRI9 -> TRI_C 0.890
## TRI10 -> TRI_C 0.992
## TRI11 -> TRI_C 0.687
## TRI12 -> TRI_C 0.761
## TRI13 -> TRI_D 0.130
## TRI14 -> TRI_D 0.705
## TRI15 -> TRI_D 0.671
## TRI16 -> TRI_D 0.963Significancia modelo segundo orden
specific_effect_significance(boot_seminr_model = boot_m_2,
from = 'TRI',
through = 'IU',
to = 'SNS',
alpha = 0.05)## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## 0.38674111 0.39751212 0.03353316 11.53309594 0.32980491
## 97.5% CI
## 0.46136011sum_boot_m_2$bootstrapped_paths ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## TRI -> IU 0.632 0.647 0.030 20.838 0.586 0.704
## IU -> SNS 0.612 0.614 0.034 17.943 0.544 0.682K.2.7. Evaluación modelo de 2do orden Reflectivo
plot(summary_m_3$reliability, title = "Fig. : Fiabilidad orden inferior")
plot(summary_m_3$paths[,1], pch = 2, col = "red", main="Betas y R^2 (Exogenos)",
xlab = "Variables", ylab = "Valores estimados", xlim = c(0,length(row.names(summary_m_3$paths))+1)
)
text(summary_m_3$paths[,1],labels = row.names(summary_m_3$paths) , pos = 4)
plot(summary_m_3$paths[,2], pch = 2, col = "red", main="Betas y R^2 (Endogenos)",
xlab = "Variables", ylab = "Valores estimados" , xlim = c(0,length(row.names(summary_m_3$paths))+1) )
text(summary_m_3$paths[,2],labels = row.names(summary_m_3$paths) , pos = 4)
summary_m_3$reliability## alpha rhoC AVE rhoA
## TRI -0.701 0.112 0.468 1.000
## IU 0.794 0.907 0.829 0.794
## SNS 0.769 0.852 0.591 0.778
## TRI_A 0.866 0.908 0.712 0.872
## TRI_B 0.816 0.878 0.643 0.838
## TRI_C 0.797 0.864 0.614 0.847
## TRI_D 0.695 0.753 0.474 0.725
##
## Alpha, rhoC, and rhoA should exceed 0.7 while AVE should exceed 0.5summary_m_3$loading## TRI IU SNS TRI_A TRI_B TRI_C TRI_D
## TRI_A 0.947 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_B 0.636 0.000 0.000 0.000 0.000 0.000 0.000
## TRI_C -0.511 -0.000 -0.000 0.000 0.000 0.000 0.000
## TRI_D -0.554 -0.000 -0.000 0.000 0.000 0.000 0.000
## IU1 0.000 0.909 0.000 0.000 0.000 0.000 0.000
## IU2 0.000 0.912 0.000 0.000 0.000 0.000 0.000
## U1 0.000 0.000 0.763 0.000 0.000 0.000 0.000
## U2 0.000 0.000 0.793 0.000 0.000 0.000 0.000
## U3 0.000 0.000 0.814 0.000 0.000 0.000 0.000
## U4 0.000 0.000 0.700 0.000 0.000 0.000 0.000
## TRI1 0.000 0.000 0.000 0.812 0.000 -0.000 -0.000
## TRI2 0.000 0.000 0.000 0.833 0.000 -0.000 -0.000
## TRI3 0.000 0.000 0.000 0.884 0.000 -0.000 -0.000
## TRI4 0.000 0.000 0.000 0.845 0.000 -0.000 -0.000
## TRI5 0.000 0.000 0.000 0.000 0.805 -0.000 -0.000
## TRI6 0.000 0.000 0.000 0.000 0.765 -0.000 -0.000
## TRI7 0.000 0.000 0.000 0.000 0.862 -0.000 -0.000
## TRI8 0.000 0.000 0.000 0.000 0.771 -0.000 -0.000
## TRI9 0.000 0.000 0.000 -0.000 -0.000 0.753 0.000
## TRI10 0.000 0.000 0.000 -0.000 -0.000 0.866 0.000
## TRI11 0.000 0.000 0.000 -0.000 -0.000 0.713 0.000
## TRI12 0.000 0.000 0.000 -0.000 -0.000 0.796 0.000
## TRI13 0.000 0.000 0.000 0.000 -0.000 0.000 0.184
## TRI14 0.000 0.000 0.000 -0.000 -0.000 0.000 0.676
## TRI15 0.000 0.000 0.000 -0.000 -0.000 0.000 0.795
## TRI16 0.000 0.000 0.000 -0.000 -0.000 0.000 0.879summary_m_3$validity$fl_criteria ## TRI IU SNS TRI_A TRI_B TRI_C TRI_D
## TRI 0.684 . . . . . .
## IU 0.624 0.910 . . . . .
## SNS 0.614 0.612 0.769 . . . .
## TRI_A 0.947 0.591 0.527 0.844 . . .
## TRI_B 0.636 0.397 0.524 0.428 0.802 . .
## TRI_C -0.511 -0.319 -0.372 -0.282 -0.502 0.784 .
## TRI_D -0.554 -0.346 -0.448 -0.337 -0.502 0.505 0.688
##
## FL Criteria table reports square root of AVE on the diagonal and construct correlations on the lower triangle.summary_m_3$validity$htmt ## TRI IU SNS TRI_A TRI_B TRI_C TRI_D
## TRI . . . . . . .
## IU 0.710 . . . . . .
## SNS 0.821 0.777 . . . . .
## TRI_A 0.836 0.707 0.636 . . . .
## TRI_B 1.018 0.481 0.655 0.489 . . .
## TRI_C 0.965 0.375 0.460 0.303 0.610 . .
## TRI_D 0.922 0.370 0.485 0.406 0.550 0.558 .summary_m_3$validity$vif_items ## TRI :
## TRI_A TRI_B TRI_C TRI_D
## 1.257 1.644 1.512 1.537
##
## IU :
## IU1 IU2
## 1.764 1.764
##
## SNS :
## U1 U2 U3 U4
## 1.459 1.501 1.708 1.422
##
## TRI_A :
## TRI1 TRI2 TRI3 TRI4
## 1.967 2.076 2.563 2.065
##
## TRI_B :
## TRI5 TRI6 TRI7 TRI8
## 1.785 1.746 1.930 1.636
##
## TRI_C :
## TRI9 TRI10 TRI11 TRI12
## 1.402 1.733 1.767 2.054
##
## TRI_D :
## TRI13 TRI14 TRI15 TRI16
## 1.239 1.663 1.690 1.375summary_m_3$validity$cross_loadings## TRI IU SNS TRI_A TRI_B TRI_C TRI_D
## TRI1 0.740 0.432 0.373 0.812 0.279 -0.164 -0.206
## TRI2 0.786 0.478 0.462 0.833 0.332 -0.227 -0.296
## TRI3 0.836 0.516 0.467 0.884 0.365 -0.275 -0.276
## TRI4 0.827 0.555 0.467 0.845 0.450 -0.271 -0.344
## TRI5 0.471 0.329 0.435 0.310 0.805 -0.353 -0.353
## TRI6 0.396 0.234 0.282 0.231 0.765 -0.295 -0.369
## TRI7 0.606 0.385 0.483 0.420 0.862 -0.506 -0.482
## TRI8 0.534 0.296 0.441 0.379 0.771 -0.420 -0.390
## TRI9 -0.360 -0.255 -0.306 -0.162 -0.405 0.753 0.460
## TRI10 -0.515 -0.331 -0.344 -0.345 -0.412 0.866 0.405
## TRI11 -0.294 -0.156 -0.212 -0.108 -0.414 0.713 0.320
## TRI12 -0.374 -0.200 -0.267 -0.194 -0.359 0.796 0.384
## TRI13 0.174 0.061 0.021 0.273 -0.037 0.083 0.184
## TRI14 -0.231 -0.193 -0.228 -0.081 -0.240 0.253 0.676
## TRI15 -0.386 -0.191 -0.312 -0.216 -0.385 0.316 0.795
## TRI16 -0.549 -0.350 -0.441 -0.346 -0.494 0.538 0.879
## TRI_A 0.947 0.591 0.527 1.000 0.428 -0.282 -0.337
## TRI_B 0.636 0.397 0.524 0.428 1.000 -0.502 -0.502
## TRI_C -0.511 -0.319 -0.372 -0.282 -0.502 1.000 0.505
## TRI_D -0.554 -0.346 -0.448 -0.337 -0.502 0.505 1.000
## IU1 0.538 0.909 0.578 0.506 0.353 -0.284 -0.298
## IU2 0.598 0.912 0.537 0.570 0.370 -0.297 -0.332
## U1 0.453 0.482 0.763 0.407 0.362 -0.225 -0.300
## U2 0.475 0.522 0.793 0.400 0.419 -0.316 -0.344
## U3 0.540 0.482 0.814 0.493 0.364 -0.305 -0.354
## U4 0.416 0.381 0.700 0.305 0.489 -0.303 -0.395K.2.8. Bootstrap modelo de 2do orden Reflectivo
boot_m_3 <- bootstrap_model(seminr_model = estimacion_model_3 , #K.2.3. b
nboot = 500, ### N° Subsamples 5000<
cores = parallel::detectCores(), #CPU cores -parallel processing
seed = 123)
plot(boot_m_3)sum_boot_m_3 <- summary(boot_m_3, alpha=0.05 ) ### Intervalo de confianza, en este caso es dos colas 90%K.2.9. Evaluación Bootstrap modelo de 2do orden Reflectivo
sum_boot_m_3$bootstrapped_weights ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## TRI_A -> TRI 0.794 0.790 0.051 15.435 0.688
## TRI_B -> TRI 0.157 0.154 0.075 2.083 0.004
## TRI_C -> TRI -0.139 -0.138 0.079 -1.758 -0.283
## TRI_D -> TRI -0.138 -0.136 0.083 -1.649 -0.299
## IU1 -> IU 0.544 0.544 0.014 39.431 0.518
## IU2 -> IU 0.554 0.555 0.013 41.414 0.530
## U1 -> SNS 0.335 0.334 0.021 15.928 0.293
## U2 -> SNS 0.362 0.362 0.023 15.701 0.320
## U3 -> SNS 0.334 0.334 0.024 14.185 0.289
## U4 -> SNS 0.264 0.264 0.019 13.845 0.226
## TRI1 -> TRI_A 0.258 0.258 0.020 12.672 0.221
## TRI2 -> TRI_A 0.285 0.286 0.018 16.086 0.251
## TRI3 -> TRI_A 0.308 0.309 0.016 19.294 0.279
## TRI4 -> TRI_A 0.332 0.330 0.020 16.736 0.292
## TRI5 -> TRI_B 0.328 0.326 0.030 10.901 0.262
## TRI6 -> TRI_B 0.233 0.231 0.031 7.441 0.169
## TRI7 -> TRI_B 0.383 0.383 0.028 13.593 0.332
## TRI8 -> TRI_B 0.295 0.297 0.039 7.575 0.218
## TRI9 -> TRI_C 0.341 0.338 0.053 6.402 0.238
## TRI10 -> TRI_C 0.442 0.449 0.049 9.066 0.358
## TRI11 -> TRI_C 0.208 0.201 0.052 3.997 0.084
## TRI12 -> TRI_C 0.267 0.265 0.040 6.650 0.181
## TRI13 -> TRI_D -0.105 -0.122 0.109 -0.958 -0.363
## TRI14 -> TRI_D 0.334 0.330 0.056 5.925 0.212
## TRI15 -> TRI_D 0.330 0.324 0.046 7.227 0.226
## TRI16 -> TRI_D 0.604 0.603 0.067 8.957 0.479
## 97.5% CI
## TRI_A -> TRI 0.885
## TRI_B -> TRI 0.299
## TRI_C -> TRI 0.026
## TRI_D -> TRI 0.033
## IU1 -> IU 0.571
## IU2 -> IU 0.584
## U1 -> SNS 0.374
## U2 -> SNS 0.410
## U3 -> SNS 0.383
## U4 -> SNS 0.299
## TRI1 -> TRI_A 0.295
## TRI2 -> TRI_A 0.318
## TRI3 -> TRI_A 0.344
## TRI4 -> TRI_A 0.371
## TRI5 -> TRI_B 0.384
## TRI6 -> TRI_B 0.287
## TRI7 -> TRI_B 0.438
## TRI8 -> TRI_B 0.370
## TRI9 -> TRI_C 0.462
## TRI10 -> TRI_C 0.551
## TRI11 -> TRI_C 0.294
## TRI12 -> TRI_C 0.337
## TRI13 -> TRI_D 0.075
## TRI14 -> TRI_D 0.430
## TRI15 -> TRI_D 0.402
## TRI16 -> TRI_D 0.746summary_m_3$validity$vif_items ## TRI :
## TRI_A TRI_B TRI_C TRI_D
## 1.257 1.644 1.512 1.537
##
## IU :
## IU1 IU2
## 1.764 1.764
##
## SNS :
## U1 U2 U3 U4
## 1.459 1.501 1.708 1.422
##
## TRI_A :
## TRI1 TRI2 TRI3 TRI4
## 1.967 2.076 2.563 2.065
##
## TRI_B :
## TRI5 TRI6 TRI7 TRI8
## 1.785 1.746 1.930 1.636
##
## TRI_C :
## TRI9 TRI10 TRI11 TRI12
## 1.402 1.733 1.767 2.054
##
## TRI_D :
## TRI13 TRI14 TRI15 TRI16
## 1.239 1.663 1.690 1.375sum_boot_m_3$bootstrapped_loadings ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## TRI_A -> TRI 0.947 0.942 0.022 43.098 0.893
## TRI_B -> TRI 0.636 0.631 0.056 11.368 0.526
## TRI_C -> TRI -0.511 -0.513 0.062 -8.297 -0.624
## TRI_D -> TRI -0.554 -0.559 0.057 -9.752 -0.662
## IU1 -> IU 0.909 0.908 0.009 96.057 0.889
## IU2 -> IU 0.912 0.912 0.009 96.936 0.892
## U1 -> SNS 0.763 0.764 0.025 30.964 0.715
## U2 -> SNS 0.793 0.793 0.021 37.454 0.751
## U3 -> SNS 0.814 0.813 0.023 35.054 0.765
## U4 -> SNS 0.700 0.701 0.031 22.359 0.636
## TRI1 -> TRI_A 0.812 0.811 0.025 31.984 0.754
## TRI2 -> TRI_A 0.833 0.833 0.021 39.232 0.788
## TRI3 -> TRI_A 0.884 0.884 0.021 43.093 0.836
## TRI4 -> TRI_A 0.845 0.845 0.019 43.361 0.805
## TRI5 -> TRI_B 0.805 0.804 0.027 29.470 0.740
## TRI6 -> TRI_B 0.765 0.764 0.036 21.410 0.686
## TRI7 -> TRI_B 0.862 0.862 0.016 53.413 0.828
## TRI8 -> TRI_B 0.771 0.773 0.037 21.082 0.695
## TRI9 -> TRI_C 0.753 0.748 0.044 16.962 0.654
## TRI10 -> TRI_C 0.866 0.867 0.023 37.847 0.819
## TRI11 -> TRI_C 0.713 0.706 0.056 12.645 0.564
## TRI12 -> TRI_C 0.796 0.790 0.038 21.080 0.704
## TRI13 -> TRI_D 0.184 0.158 0.146 1.261 -0.136
## TRI14 -> TRI_D 0.676 0.660 0.084 8.040 0.446
## TRI15 -> TRI_D 0.795 0.780 0.050 15.794 0.663
## TRI16 -> TRI_D 0.879 0.873 0.029 30.111 0.811
## 97.5% CI
## TRI_A -> TRI 0.977
## TRI_B -> TRI 0.731
## TRI_C -> TRI -0.389
## TRI_D -> TRI -0.443
## IU1 -> IU 0.925
## IU2 -> IU 0.928
## U1 -> SNS 0.808
## U2 -> SNS 0.833
## U3 -> SNS 0.854
## U4 -> SNS 0.756
## TRI1 -> TRI_A 0.855
## TRI2 -> TRI_A 0.868
## TRI3 -> TRI_A 0.918
## TRI4 -> TRI_A 0.879
## TRI5 -> TRI_B 0.852
## TRI6 -> TRI_B 0.823
## TRI7 -> TRI_B 0.891
## TRI8 -> TRI_B 0.831
## TRI9 -> TRI_C 0.823
## TRI10 -> TRI_C 0.908
## TRI11 -> TRI_C 0.793
## TRI12 -> TRI_C 0.851
## TRI13 -> TRI_D 0.424
## TRI14 -> TRI_D 0.787
## TRI15 -> TRI_D 0.846
## TRI16 -> TRI_D 0.926Significancia modelo segundo orden
specific_effect_significance(boot_seminr_model = boot_m_3,
from = 'TRI',
through = 'IU',
to = 'SNS',
alpha = 0.05)## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## 0.3819306 0.3888027 0.0319770 11.9439172 0.3243270
## 97.5% CI
## 0.4515041sum_boot_m_3$bootstrapped_paths ## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI 97.5% CI
## TRI -> IU 0.624 0.632 0.028 22.111 0.572 0.685
## IU -> SNS 0.612 0.614 0.034 17.943 0.544 0.682L. Analisis Pathmox
L.1. Defición modelo
Definir modelo usando laavan syntax.
cSmodel <- "
# Structural model
SNS ~ IU + FC + HA
IU ~ FC + HA + SI + HM + PE + EE
#modelo de medida
PE =~ PE1 + PE2 + PE3 + PE4
EE =~ EE1 + EE2 + EE3
SI =~ SI1 + SI2 + SI3 + SI4
FC =~ FC1 + FC2 + FC3
HM =~ HM1 + HM2 + HM3
HA =~ HA1 + HA2 + HA3 + HA4 + HA5
IU =~ IU1 + IU2
SNS =~ U1 + U2+ U3 + U4
"L.2. Analisis con cSEM
est_model <- csem(.data = pls_data2, .model = cSmodel)
bootstrap<- csem(.data = pls_data2, .model = cSmodel, .resample_method = "bootstrap", .R = 1000)
#summarize(bootstrap)
#summarize(est_model)
#valides <- assess(est_model)
#infer(est_model)
#predict(est_model)
#verify(est_model)L.3. Configurando las variables
Variables deben estar como factor
pls_data2$GENDER2= as.factor(pls_data2$GENDER)
pls_data2$EDU2= as.factor(pls_data2$EDU)
pls_data2$RETIRED2= as.factor(pls_data2$RETIRED)
pls_data2$WSTATUS2= as.factor(pls_data2$WSTATUS)
pls_data2$GENERATION2= as.factor(pls_data2$GENERATION)
pls_data2$REGION2= as.factor(pls_data2$REGION)
pls_data2$EXP2= as.factor(pls_data2$EXP)
pls_data2$SOC2= as.factor(pls_data2$SOC)pls_data2$TRI_A = pls_data2$TRI1 + pls_data2$TRI2 + pls_data2$TRI3 + pls_data2$TRI4
pls_data2$TRI_B = pls_data2$TRI5 + pls_data2$TRI6 + pls_data2$TRI7 + pls_data2$TRI8
pls_data2$TRI_C = pls_data2$TRI9 + pls_data2$TRI10 + pls_data2$TRI11 + pls_data2$TRI12
pls_data2$TRI_D = pls_data2$TRI13 + pls_data2$TRI14 + pls_data2$TRI15 + pls_data2$TRI16
pls_data2$TRI_T <- ifelse(pls_data2$TRI_B <= pls_data2$TRI_A & pls_data2$TRI_C <= pls_data2$TRI_A
& pls_data2$TRI_D <= pls_data2$TRI_A, 1,
ifelse (pls_data2$TRI_A <= pls_data2$TRI_B & pls_data2$TRI_C <= pls_data2$TRI_B
& pls_data2$TRI_D <= pls_data2$TRI_B, 2,
ifelse (pls_data2$TRI_A < pls_data2$TRI_C & pls_data2$TRI_B <= pls_data2$TRI_C
& pls_data2$TRI_D <= pls_data2$TRI_C, 3, 4)))
pls_data2$TRI_T3= pls_data2$TRI_T
pls_data2$TRI_T2= as.factor(pls_data2$TRI_T)Genero grupo de categoricas
categoricas2 <- c( #"EXP2",
"EDU2", "SOC2" , "WSTATUS2", "RETIRED2" ,"GENDER2" , "GENERATION2", "REGION2", "TRI_T2")Conjunto de datos con categoricas
CSIcatvar <- pls_data2[, categoricas2]L.4. Generacion modelo y resultado
Ejecutar analisis Phatmox (Lamberti et al., 2016; 2017)
csi.pathmox = pls.pathmox(
.model = cSmodel ,
.data = pls_data2,
.catvar= CSIcatvar, ## Variables categoricas a ser utilizadas
# .scheme= 'centroid', 'factorial', 'path' defecto Tupo de esquema de ponderación interna
.size = 0.10, #minimo de observaciones en porcentaje
.size_candidate = 50, #minimo de observaciones en cantidad por defecto es 50
# .consistent = TRUE, #Default
.alpha = 0.05, ### umbral mínimo de importancia defecto 0.05
.deep = 5 ### Maxima profundidad de los arboles
) ##
## PLS-SEM PATHMOX ANALYSIS
##
## ---------------------------------------------
## Info parameters algorithm
## parameters algorithm value
## 1 threshold signif. 0.05
## 2 node size limit(%) 0.10
## 3 tree depth level 5.00
##
## ---------------------------------------------
## Info segmentation variables
## nlevels ordered treatment
## EDU2 4 FALSE nominal
## SOC2 5 FALSE nominal
## WSTATUS2 2 FALSE binary
## RETIRED2 2 FALSE binary
## GENDER2 2 FALSE binary
## GENERATION2 3 FALSE nominal
## REGION2 2 FALSE binary
## TRI_T2 4 FALSE nominalplot(csi.pathmox) Ranking de importancia de las variables
variables <-csi.pathmox[["var_imp"]][["variable"]]
ranking <- csi.pathmox[["var_imp"]][["ranking"]]
barplot(ranking, main = "Ranking de importancia de las variables",
xlab = "variables",
ylab = "Valor",
col = rainbow(10),
names.arg = variables
) 
Resultados
summary(csi.pathmox)##
## PLS-SEM PATHMOX ANALYSIS
##
## ---------------------------------------------
## Info parameters algorithm:
## parameters algorithm value
## 1 threshold signif 0.05
## 2 node size limit(%) 0.10
## 3 tree depth level 5.00
## ---------------------------------------------
## Info tree:
## parameters tree value
## 1 deep tree 2
## 2 number terminal nodes 4
## ---------------------------------------------
## Info nodes:
## node parent depth type terminal size % variable category
## 1 1 0 0 root no 383 100.00 <NA> <NA>
## 2 2 1 1 node no 215 56.14 TRI_T2 1/2
## 3 3 1 1 node no 168 43.86 TRI_T2 3/4
## 4 4 2 2 least yes 99 25.85 WSTATUS2 N
## 5 5 2 2 least yes 116 30.29 WSTATUS2 Y
## 6 6 3 2 least yes 73 19.06 EDU2 4
## 7 7 3 2 least yes 95 24.80 EDU2 1/2/3
## ---------------------------------------------
## Info splits:
##
## Variable:
## node variable g1.mod g2.mod
## 1 1 TRI_T2 1/2 3/4
## 2 2 WSTATUS2 N Y
## 3 3 EDU2 4 1/2/3
##
## Info F-global test results (global differences):
## node F value Pr(>F)
## [1,] 1 2.8561 0.0011 **
## [2,] 2 2.8227 0.0014 **
## [3,] 3 2.0571 0.0232 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Info F-coefficient test results (coefficent differences) :
##
## Node 1 :
## F value Pr(>F)
## FC -> IU 0.7652 0.3820
## HA -> IU 9.6143 0.0020 **
## SI -> IU 0.3196 0.5720
## HM -> IU 0.9399 0.3326
## PE -> IU 2.1189 0.1459
## EE -> IU 1.9001 0.1685
## FC -> SNS 0.3249 0.5688
## HA -> SNS 1.0578 0.3041
## IU -> SNS 4.0827 0.0437 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Node 2 :
## F value Pr(>F)
## FC -> IU 1.6655 0.1976
## HA -> IU 1.6850 0.1950
## SI -> IU 8.1413 0.0045 **
## HM -> IU 2.7871 0.0958 .
## PE -> IU 4.3042 0.0386 *
## EE -> IU 0.1956 0.6586
## FC -> SNS 1.4242 0.2334
## HA -> SNS 5.5389 0.0191 *
## IU -> SNS 0.6483 0.4212
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Node 3 :
## F value Pr(>F)
## FC -> IU 5.1724 0.0236 *
## HA -> IU 1.5674 0.2115
## SI -> IU 0.2740 0.6010
## HM -> IU 3.3059 0.0700 .
## PE -> IU 4.5573 0.0336 *
## EE -> IU 2.2010 0.1389
## FC -> SNS 5.3666 0.0212 *
## HA -> SNS 1.9446 0.1642
## IU -> SNS 0.7523 0.3864
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## ---------------------------------------------
## Info variable importance ranking:
## variable ranking
## 1 EDU2 0.15807663
## 3 GENERATION2 0.15239267
## 4 REGION2 0.14128688
## 7 TRI_T2 0.13532941
## 8 WSTATUS2 0.12649435
## 6 SOC2 0.10452209
## 2 GENDER2 0.09941366
## 5 RETIRED2 0.08248430
##
## ---------------------------------------------
## Info terminal nodes PLS-SEM models (path coeff. & R^2):
## node 4 node 5 node 6 node 7
## FC->IU 0.0961 -0.0507 0.2716 -0.0113
## HA->IU -0.0431 0.1008 0.3834 0.0370
## SI->IU 0.1656 -0.0233 0.3728 0.3557
## HM->IU 0.6981 0.3756 0.3845 0.6388
## PE->IU 0.4734 0.1656 0.1932 0.3033
## EE->IU 0.1068 0.3919 -0.0116 0.2746
## FC->SNS 0.0698 0.3600 0.4437 0.1508
## HA->SNS -0.0553 -0.0193 -0.2446 -0.0641
## IU->SNS 0.2289 0.3259 0.1295 0.0542
## R^2 IU 0.5540 0.5337 0.5871 0.5474
## R^2 SNS 0.6710 0.4484 0.4945 0.4799